Previous studies have demonstrated that acute systemic inflammation after surgery or infection is associated with a transient increase in the risk of vascular events. This suggests that vascular risk is not stable but fluctuates within short periods in response to inflammatory stimuli. While an association between respiratory tract infection and vascular events is well-documented, the effects of other acute infections and inflammatory stimuli are less certain. The principal aim of this project was to investigate further the role of acute inflammation and infection in vascular disease, employing the unique opportunities offered by electronic health record databases. Two large observational studies were undertaken. First, the self-controlled case series method and Medicaid claims data from the United States were used to examine the short-term effects of invasive dental treatment, a novel acute inflammatory model, on the risk of vascular events. Second, a matched case-control study using primary care data from the United Kingdom General Practice Research Database investigated the role of acute maternal infection in the development of pre-eclampsia – a vascular disorder of pregnancy. The case series analysis of 1152 adults with a vascular event demonstrated a transiently increased vascular event rate in the four weeks after invasive dental treatment relative to unexposed time periods. The analysis of 1533 pre-eclampsia cases and 14236 controls who had completed a pregnancy without pre-eclampsia revealed an increased risk of pre-eclampsia associated with antibiotic prescriptions and urinary infection, but not respiratory infection, during pregnancy. The findings suggest that exposures sufficient to produce an acute inflammatory response may play an important role in the occurrence of vascular outcomes. Future research on the effects of other acute inflammatory triggers and the mediating mechanisms involved should help establish a clearer role for acute inflammation and infection in vascular disease and inform preventative strategies during periods of increased vascular risk.

Congenital heart disease occurs in 0.8-1% of children. A proportion of these patients will require interventional treatment in the form of cardiac catheterisation or surgery, whilst some will only require medical therapy. Echocardiography is the investigation of choice for diagnosis and follow-up of these patients, whilst magnetic resonance imaging (MRI) is now increasingly used in most centres to assess complex congenital cases and to answer specific questions, not possible to address with echocardiography. The ability to obtain anatomical together with quantitative physiological information, such as cardiac function and flows in one examination has led to more detailed assessment and analysis of congenital heart defects and continues to improve our understanding about congenital heart disease and its treatment options. Cardiac MRI scans have now mostly replaced cardiac catheterisation procedures, routinely performed up to a few years ago to aid diagnosis. In parallel, a new form of hybrid catheterisation has emerged by combining MRI with simultaneous pressure measurement in different cardiac chambers and vascular structures. The combination of X-Ray and MRI guided (XMR) catheterisations can address clinical questions, such as estimation of pulmonary vascular resistance and cardiac output response to stress accurately, without the limitation of haemodynamic assumptions during calculations. Further to more accurate physiological information, XMR catheterisation can also offer detailed anatomical information of structures not well seen on echocardiography or MRI alone, and most importantly limit or eliminate the radiation exposure to patient that have been repetitively exposed to X-Ray radiation. The purpose of the research work presented was two-fold: a. to extend the potentials of XMR-guided cardiac catheterisations, particularly in the area of pre-operative and post-interventional evaluation of congenital heart disease and b. to advance our already established programme of solely MR-guided diagnostic cardiac catheterisations in order to materialise the first-in-man solely MR-guided therapeutic cardiac catheterisations. In this way, transcatheter interventional procedures were aimed to be performed in the MR scanner in a similar fashion and with similar equipment to the ones used in the traditional catheterisation suite, without the exposure to ionising radiation. The work towards materialisation of this idea and the world-first clinical trial on solely MRI-guided cardiac interventions for congenital heart disease are being presented in this thesis.

Fetal haemoglobin levels have a major impact on the severity of sickle cell disease (SCD) and Beta-Thalassaemia. Previous studies by our group shows that three loci – the HBS1L-MYB intergenic region (HMIP) on chromosome 6q23, BCL11A on chromosome 2p16 and the β globin cluster on chromosome 11 accounts for up to 50 % of the variation in HbF levels in non-anaemic Europeans. Recent work by our group indicated the presence of regulatory elements in HBS1L-MYB intergenic region (HMIP) evidenced by GATA-1 binding coinciding with strong histone acetylation, RNA polymerase II activity, and erythroid specific DNase I hypersensitive sites. Genetic studies and resequencing of two key individuals from the family that led to discovery of 6q QTL revealed new SNPs in the HBS1L-MYB intergenic region. In silico predictions suggested that a couple of these SNPs altered binding of key transcription factors. We proposed that differential binding of these transcription factors mediated through these SNPs underlie the association with HbF control. Electrophoretic Mobility Shift Assays (EMSAs) and Chromatin immunoprecipitation followed by real time PCR (ChIP-qPCR) and HaploChIP showed that many of the genetically associated variants altered the binding of key transcription factors such as GATA-1, KLF1 and MYB that have crucial roles in globin regulation as well as regulation of erythropoiesis. We have also compared ex-vivo erythroid cell culture kinetics and expression profiles of individuals who are homozygous for the presence or absence of the allelic variant associated with HbF. MYB expression and maturation rate of erythroblasts were noticeably different whereas expression of HBS1L the other flanking gene was not different. We could also demonstrate unequal expression of MYB in individuals heterozygous for the HMIP variants but the direction of the effect should be further investigated. These results further support the regulatory role of HBS1L-MYB intergenic polymorphisms on HbF and erythropoietic kinetics supporting our hypothesis that the effect of the intergenic SNPs is indirect via the flanking MYB gene.

This thesis describes the development and evaluation of a self-management intervention for people admitted to hospital with heart failure due to left ventricular systolic dysfunction. The intervention was delivered by a heart failure nurse specialist, following optimisation of in-patient medical treatment. It consisted of two sessions during the hospital stay, a home visit and follow-up telephone call. The content of the intervention and techniques for its delivery were detailed in a manual. The intervention was evaluated in a single-blind randomised controlled trial. The primary outcomes were the number of readmissions to hospital and duration of hospital stay in the 90 days after discharge from the index admission. Other outcomes included readmissions to hospital and duration of hospital stay in the 12 months after discharge, mortality, quality of life and psychological well-being. Process variables included self- management behaviours and cognitions. One hundred and sixty-five patients were recruited to the study and they were followed up for a period of 12 months. The intervention had a significant effect on aspects of self-management behaviour but did not demonstrate a significant effect on readmission to hospital or mortality in the total sample. However, a post-hoc sub-group analysis identified a significant impact of the intervention in those patients who were newly diagnosed with heart failure. Newly diagnosed patients in the intervention group had fewer admissions to hospital for heart failure and spent fewer days in hospital for heart failure during the 12 month follow-up period than those in the control group. The implications of the findings for future research and intervention development are discussed.

Heart failure (HF) and sudden cardiac death (SCD) in congenital heart disease (CHD) is prevalent and can relate to abnormal right ventricular (RV) physiology and abnormalities of QRS duration, and QRS, JT and QT dispersion (d). Characterising disease and identifying factors that may predict adverse outcome in those with either a subpulmonary or subsystemic RV, as well as investigating potential avenues to ameliorate abnormal RV physiology is necessary to improve outcomes in this young population. I undertook several studies during the course of this Thesis to examine and further understand these two separate physiological substrates: In the first I studied the effect of isolated percutaneous (PPVI) pulmonary valve implantation on surface ECG parameters. PPVI represents a pure model of RV mechanical and electrophysiological changes post replacement as compared to surgical replacement: Ninety nine PPVI procedures in patients with CHD (aged 23.1±10 yrs) were studied pre, post and 1-year following PPVI with serial ECG’s and echocardiography/ magnetic resonance imaging (CMR). 43% had pulmonary stenosis, 27% pulmonary regurgitation (PR) and 29% mixed lesions. In those with predominantly PR (n=26), QRS duration decreased significantly (135±27 to 128±29ms; p=0.007). However, in the total cohort no significant change in QRS duration at 1 year was observed (137±29 to 134±29ms). QTc, QRSd, QTd and JTd all significantly reduced at 1 yr (p≤0.001). RV EDV correlated with pre-procedure QRS duration (r=0.34; p<0.002) but there was no correlation after PPVI. This is the first study to report electrical remodelling following isolated PPVI and it confirms that reductions in QRS duration occur post PPVI in PR, as reported for equivalent surgical cohorts. Further, increased homogeneity of repolarisation, in combination with improved conduction, may reduce arrhythmic events in congenital cardiac patients with pulmonary valvular disease. My second study sought to create an epicardial electroanatomic map of the RV and then apply post-operative targeted single and dual site RV temporary pacing with measurement of haemodynamic parameters. I wished to determine the potential role of cardiac resynchronization therapy (CRT) in the setting of RV dysfunction as little is known regarding the potential benefits of CRT in this setting. Sixteen adults (age=32±8 yrs, 6M; 10F) with right bundle branch block (RBBB) and repaired tetralogy of Fallot (ToF; n=8) or corrected congenital pulmonary stenosis (n=8) undergoing surgical pulmonary valve replacement (PVR) for PR underwent intra-operative epicardial RV mapping and haemodynamic assessment of random pacing configurations including site of latest RV activation. I found that the commonest site of latest activation was the RV free wall & dual chamber (DDD) pacing here, alone or combined with RV apical pacing, resulted in significant increases in cardiac output (CO) vs AAI pacing (p<0.01 all measures). DDD RV alternative site pacing significantly improved CO by 16 % vs AAI, and 8.5% versus DDD RV apical pacing (p=0.02). Single site RV pacing targeted to the region of latest activation in patients with RBBB undergoing PVR thus induces acute improvements in haemodynamics and implies that targeted pacing in such patients has therapeutic potential both post-operatively and in the long term. QRS duration is a strong predictor of survival in acquired left ventricular dysfunction, but equivalent data in those with a systemic RV is lacking. My next studies investigated not only the relationship between ECG parameters, arrhythmia burden and outcome in adults with transposition of the great arteries (TGA) late after atrial switch repair, but also the interrelationships between various HF markers in this cohort. Adults with Senning or Mustard palliation of TGA under follow up at a dedicated congenital HF clinic and 13 similar adults who suffered a cardiac death were included for study. Patients were subdivided by arrhythmic history, surgical intervention and death. Assessment included symptom assessment, venous blood sampling for circulating N-terminal pro brain natriuretic peptide (NT-proBNP) levels, measurement of surface ECG and CMR for the assessment of RV systolic function and determination of indexed RV volumes. I found that QRS duration (p=0.0003) and QTc interval (p=0.0009) increase significantly with changing arrhythmia subtype, and that both QRS and QTc were independently associated with increased risk of death: for 1ms increase in QRS HR 15 [95% CI 3.3-68.6] and for QTc HR 10.7 [95% CI 2.3-49] (p<0.0001 for both). QRS >104ms and QTc >406ms had a sensitivity/specificity for predicting death of 96%/66% and 96%/56% respectively. Two year mortality was 36% when QRS<104ms and 88% when >104ms (p<0.0001 for difference). Further, compared to those with uncomplicated surgery, patients with complex surgical history had higher NT-proBNP levels (55±26 vs 20±35pmol/L; P=0.002) and longer QRS duration (116±28ms vs 89±11ms; P=0.0004) whilst showing no difference in NYHA class and RV function. There was a significant relationship between diastolic and systolic RV volumes and both NT-proBNP levels (r=0.43, P=0.01; r=0.53, P=0.001 respectively) and QRS duration (r=0.47, P=0.004; r=0.53, P=0.001 respectively). These findings suggest that QRS width and corrected QTc interval on surface ECG are associated with increased risk of death in adults late after atrial switch repair of TGA. Given that a QRS of only 104ms defines a high risk population, careful examination of the ECG is desirable in all patients and therapy to reduce risk attempted. Further, together with these simple surface ECG parameters, circulating NT-proBNP levels constitute safe, cost effective and widely available surrogate markers of systemic RV function and provide additional information on heart failure status. Both measures hold promise as prognostic markers and their association with long-term outcome should be determined. Lastly, I examined the mechanisms of late RV failure and studied their relationship to subjective quality of life assessment as this are poorly characterised. Equilibrium Contrast CMR imaging was used to quantify extracellular volume (ECV) in the septum and RV free wall of adults presenting to a specialist clinic late after atrial redirection surgery for TGA. These were compared to age and sex matched healthy volunteers. Patients were also assessed with a standardised CMR protocol, NT-proBNP and surface ECG measurement, and cardiopulmonary exercise (CPEX) testing. Patients also completed a Minnesota Living With Heart Failure Questionnaire (MLHFQ) self assessment. I determined that mean septal ECV was significantly higher in patients than controls (0.254±0.036, vs 0.230±0.032; p=0.03). NT-proBNP positively related to septal ECV (p=0.04; r=0.55) but chronotropic index (CI) during CPEX testing negatively related to ECV (p=0.04; r=-0.58). No relationship was seen with other CMR or CPEX parameters. Median MLHFQ score was 6(2-19), median NT-pro BNP 24 (16-43) and mean peak VO2 24±7mL/kg/min. There was a significant positive correlation between MLHFQ score and NT-proBNP (p=0.001, r=0.34) and a significant negative correlation with peak VO2 (p=0.001, r=0.49. ). Septal interstitial expansion is seen in adults late after atrial redirection surgery for TGA. It correlates well with NT-proBNP and CI and may have a role in the development of RV systolic impairment. The MLHFQ correlates highly with NT-proBNP and exercise capacity in patients with systemic RV impairment. The ability of the MLHFQ in predicting HF events and prognosis in adults with CHD needs further evaluation.

This thesis describes the evaluation of a mobile telehealth intervention for people with diabetes. In line with the gaps in the literature identified in a systematic review of mobile telehealth interventions in diabetes, the aims of this thesis were to test the effectiveness of a mobile telehealth intervention, to examine the mechanisms of action of the intervention, to identify predictors of benefit and of telehealth usage, to explore patients’ perspectives on telehealth and to assess intervention fidelity. A randomized controlled trial comparing standard care to standard care supplemented with mobile telehealth was conducted and included 81 participants with insulin-requiring and poorly controlled diabetes. Participants completed assessments at baseline, 3 months and 9 months. Outcomes included glycaemic control (HbA1c), self-management behaviours, quality of life and health care utilization. Efficacy beliefs and personal models of diabetes were assessed as process variables. In addition, mobile telehealth participants completed assessments of acceptability of telehealth, perceived adequacy of training and self-efficacy to use telehealth. Semi-structured interviews conducted with intervention group participants having completed the study explored patients’ perceptions and experiences using mobile telehealth. Results indicated the intervention had no effect on study outcomes. Efficacy beliefs in the intervention group were found to improve significantly compared to the control group and there was some evidence in mediation analyses to suggest these changes mediated improvements in study outcomes. Some evidence also suggested that improvements in efficacy beliefs were predicted by clinical variables. In terms of patients’ usage of telehealth, weekly frequency of self-monitoring of blood glucose and dietary behaviours at baseline were the strongest predictors, but results also suggested that acceptability, self-efficacy to use telehealth, perceived adequacy of training, and the number of contacts with telehealth providers were related to telehealth usage. The qualitative interviews highlighted features of the mobile telehealth equipment that were particularly popular. They also provided valuable information on the perceived effects of mobile telehealth and on key areas of improvement for the successful implementation of telehealth in routine practice. Data collected on the delivery of the intervention suggested it was unlikely to have been delivered faithfully.

Background: Natural killer (NK) cells are cytotoxic effectors providing a first line of defence against viruses and tumours. NK cells can be isolated from peripheral blood (PB) or cord blood (CB) for cancer immunotherapy. However, it was shown by our group and others, that CB NK cells express higher NKG2A and less killer immunoglobulin like receptors (KIRs) than PB NK cells indicating an immature phenotype. Also, CB NK cells require high doses of interleukin (IL)-2 for proliferation and activation. It was also shown that resting CB NK cells are poorly cytotoxic and produce less IFN-γ than PB NK cells after stimulation with IL-2. Hypothesis and aims: CB NK cells have an immature phenotype and could mediate different role in neonatal immunity than adult PB NK cells. Hence, the aim of this study was to explore whether differential mechanisms of activation exist between PB and CB NK cells. Methods: PB samples from healthy volunteers and CB samples were obtained with prior written consent and ethical approval from Anthony Nolan. Purified PB and CB NK cells were stimulated with cytokines including IL-2, IL-12, IL-15, IL-18, individually or in combination. Thereafter, comparative analysis was performed on their phenotype, signalling, proliferation, cytotoxicity, cytokine secretion and chemotaxis post-cytokine stimulation. Results: My results show that CB NK cells responded less to IL-2 activation than PB NK cells, which correlated with lower levels of IL-2 receptors and decreased phosphorylation of STAT5 pathway. CB NK cells activated with IL-15+IL-2 showed enhanced cytotoxicity while activation with IL-15+IL-18 promoted maximal proliferation, higher IFN-γ and TNF-α secretion. In contrast, optimal activation of PB NK cells was achieved by IL-2 stimulation. Interestingly, CB NK cells secreted substantial IL-8 concentrations following cytokine stimulation. IL-12 or IL-18 stimulation induced L- selectin expression on CB NK cells and promoted NK cell migration towards chemokines that induce homing to lymph nodes. I also generated long-lived memory- like NK cells from PB and CB using cytokines whereby IL-12+IL-15+IL-18 pre- activation led to substantial IFN-! production. Conclusions: CB NK cells are fully functional upon activation with IL-15+IL-2 or IL-15+IL-18 rather than IL-2 thereby providing a basis for activation of NK cells derived from different sources that may be utilised for future NK cell-based therapeutic purposes. In addition, CB NK cell cytokine secretion profile is suggestive for a role of neonatal NK cells in providing protective immunity against bacterial infections.

Systemic amyloidosis is the exemplar infiltrative, extracellular disease. Although it is a multi-organ disorder, cardiac involvement drives prognosis. Survival is worst in the AL amyloidosis subtype. It can affect any age and any race. There is no direct test for amyloid burden and there is no treatment for amyloidosis, there is only treatment for the underlying condition. Earlier diagnosis permits prompt treatment and improves survival. A number of imaging modalities exist to non-invasively detect cardiac disease but all have limitations. Cardiovascular magnetic resonance (CMR) with late gadolinium enhancement (LGE) imaging provides the highest sensitivity for early detection. However, this also has its shortcomings. There is currently no non-invasive method of directly measuring amyloid burden in the extracellular space. New therapies are pending – but their development needs new surrogate endpoints and new tests are therefore desperately needed. T1 mapping permits tissue abnormalities to be directly visualised in a simple scan – the colour changes being instantly recognisable, either before contrast (pre contrast or native T1 mapping) or after, when the myocardial extracellular volume (ECV) can be measured. In a collaboration between the National Amyloidosis Centre and the Heart Hospital, I explored the possibility and potential that T1 mapping might measure cardiac (and other organ) involvement in systemic amyloidosis using EQ-MRI. In early clinical exploration in systemic AL amyloid, I showed that native myocardial T1 was elevated in cardiac amyloidosis and tracked disease, particularly early disease. Mean pre contrast myocardial T1 as measured by ShMOLLI was higher in patients at 1086 ± 90msec, compared to healthy volunteers of 958 ± 20msec (P<0.001). Myocardial T1 times showed a stepwise elevation as the probability for cardiac involvement increased: 1009 ± 31msec without cardiac involvement, 1048 ± 48msec with possible cardiac involvement, 1140 ± 61msec with definite cardiac involvement (P<0.001). Using contrast to measure the ECV, I was able to non-invasively and directly measure the amyloid burden in the heart for the first time. Mean cardiac ECV was greater in patients compared to healthy volunteers with a wider range (0.44 ± 0.12 vs 0.25 ± 0.02, P<0.001) and tracked pre-test probability of cardiac involvement by conventional parameters (P<0.001). ECV also tracked conventional measures of disease severity and correlated with survival with a median ECV of 0.45 being the best model for assessing survival: HR 3.84 (1.53 – 9.61), P=0.004. I demonstrated good reproducibility of the technique with an ICC of > 0.9 for both the FLASH IR and ShMOLLI techniques of T1 mapping and good agreement of ECV derived from both techniques. In pilot studies, I also demonstrated by serial scanning that changes (including regression) over time could be measured. In other organs, I showed that the amyloid burden could be measured and was higher in amyloidosis compared to healthy volunteer: ECV 0.32 vs 0.29 (P<0.001) for liver, 0.39 vs 0.34 (P<0.001) for spleen and 0.16 vs 0.09 (P<0.001) for skeletal muscle. These ECVs also tracked current conventional measures of disease severity by nuclear scintigraphy. These results demonstrate that the interstitial volume in patients with systemic AL amyloidosis can be measured non invasively in the heart, liver, spleen and skeletal muscle and that this correlates with existing markers of disease and survival. Pre contrast myocardial T1 was a good alternative measure for the heart. In conclusion, the work in this thesis has enabled a deeper understanding of cardiac amyloidosis, disease processes and stages. It has pioneered a new prognostic marker that is also able to identify some patients with cardiac involvement that were previously unrecognised. Novel subtypes are now recognised (e.g. cardiac amyloidosis with no LVH) and it has also allowed direct quantification of the liver and spleen. ECV is a new and powerful biomarker that has already been adopted by industry allowing development of new therapies and providing hope that an end to the scourge of this disease is near.

Liver cirrhosis is associated with cardiovascular dysfunction including decreased heart rate variability and impaired acceleration of the heart rate in response to sympathetic activation (chronotropic incompetence). In this thesis, the hypothesis that increased formation of reactive nitrogen species in cirrhosis causes nitration or S-nitrosation of cardiac proteins and leads to impaired chronotropic function was assessed in an experimental model of cirrhosis. Cardiac chronotropic responsiveness to fi-adrenergic stimulation was assessed in vitro using spontaneous beating rat isolated atria. A novel mass spectrometric method was developed for dynamic assessment of nitration reactions based on the nitration of deuterium-labelled pra-hydroxyphenyl acetic acid. Nitration of cardiac proteins was measured by mass spectrometry and located by immunogold electron microscopy. Marked impairment of chronotropic responses of isolated atria to isoproterenol was observed in rats with cirrhosis, which normalized after the administration of N-acetylcysteine (a scavenger of reactive oxygen and nitrogen species) or L-NAME (a nitric oxide synthase inhibitor). The levels of protein-bound nitrotyrosine in atrial tissue increased from 16 1 to 23 3 pg/g tyrosine in rats with cirrhosis, and decreased to 15 1 and 17 1 pg/g after treatment with L-NAME and N-acetylcysteine, respectively (P<0.05). Immunogold electron microscopy demonstrated increased nitration of filaments and mitochondria in the atria of rats with cirrhosis. A chemiluminescence- based method was developed to stabilise and measure S-nitrosothiols in tissues. There was no difference in cardiac S-nitrosothiols following induction of cirrhosis, and neither N-acetylcysteine nor L-NAME had any effect on the cardiac levels of S-nitrosothiols. Autonomic regulation of cardiac function was assessed by analysis of heart rate variability in anesthetized rats using Fast Fourier Transformation. Heart rate variability analysis showed impaired sympathovagal balance towards increase of cardiac sympathetic activity in rats with cirrhosis (P<0.05). However there was no change in the sympathovagal balance following N-acetylcysteine or L-NAME adminstration in cirrhotic rats. In conclusion, abnormal cardiac chronotropic function in cirrhosis is associated with increased nitration of cardiac proteins. Two independent treatments (N-acetylcysteine and L-NAME) that decrease nitration of cardiac proteins led to normalization of cardiac responses. Nitration of critical proteins in cardiac tissue may lead to abnormal cardiac function.

Haematopoiesis is controlled by complex signal transduction pathways and transcription regulators which may become dysregulated in acute myeloid leukaemia (AML). Activating mutations in FLT3 and c-KIT receptor tyrosine kinases (RTKs) are commonly found in AML and can impact on prognosis. Different types of FLT3 mutations are known to have distinct biological activities and prognostic implications. The presence of c-KIT mutations has been shown to increase the risk of relapse, but there has been no direct comparison of the biological activity of AML specific c-KIT mutations occurring in different receptor domains. In addition to prognostic information, RTKs are attractive potential targets for therapy using small molecule inhibitors. To evaluate their biological activity and response to targeted therapies, human UT-7 cells were transduced with wild-type and mutant c-KIT isoforms - c-KIT-Δ417-419>Y (extracellular domain [ECD]), c-KIT-L576P (juxtamembrane domain [JMD]), c-KIT-D816V and c-KIT-N822K (both in the activating loop domain [ALD). There were intrinsic differences in signal strength between the mutants examined - only c-KIT-Δ417-419>Y and c-KIT-D816V expressing cells had detectable constitutive c-KIT activation and showed ligand-independent growth. The response of transduced UT-7 cells to c-KIT inhibitors was assessed by treating the cells with dasatinib and masitinib. Cells expressing ALD c-KIT mutations were more resistant to dasatinib or masitinib-mediated cell killing in comparison to cells expressing c-KIT mutations in the ECD and JMD. Western blotting revealed that although c-KIT phosphorylation was potently inhibited, there was residual mTOR and/or PI3K/AKT activation in these cells. The resistance to dasatinib observed in c-KIT-D816V or c-KIT-N822K expressing cells could be overcome by co-blockade of c-KIT and PI3K/mTOR and blockade of c-KIT and PI3K/mTOR was synergistic in all c-KIT mutant cell lines at inducing cell death. Blockade of FLT3 and PI3K/mTOR in FLT3-ITD AML cell lines also showed similar results. During the screening of AML cell lines for c-KIT and FLT3 mutations a novel FLT3-T820N point mutation was identified in the ME-1 cell line. Expression of FLT3-T820N in 32D cells constitutively activated FLT3 and conferred ligand-independent growth. 32D FLT3-T820N cells were most sensitive to the FLT3 inhibitor AC220 with regard to growth inhibition, cell killing and decreased phosphorylation of FLT3 compared to FLT3-WT and FLT3-D835Y expressing cells. This work highlights the differences in biological outcomes and TK inhibitor-sensitivity between different RTK mutants found in AML and shows that simultaneous blockade of RTKs and PI3K/mTOR may provide a novel therapeutic approach for specific AML subtypes.