Haematological malignancies and their treatment lead to prolonged periods of severe thrombocytopenia (platelet count ≤ 50 x 10⁹/l). Despite the use of prophylactic platelet transfusions, haemorrhage remains an important complication during this thrombocytopenic period. Within a 30 day period up to 70% of patients have clinically significant haemorrhage (World Health Organization (WHO) grade 2 or above bleeding) and up to 10% have severe or life-threatening haemorrhage (WHO grade 3 or 4 bleeding). Hence our current management of these patients to prevent haemorrhage is sub-optimal. The aim of this thesis was to identify clinical and laboratory factors that may predict the risk of haemorrhage in patients with haematological malignancies and severe thrombocytopenia. This was achieved via several different study designs and assessed the effect of clinical and laboratory factors on any or clinically significant haemorrhage and their effect on intracranial haemorrhage. This thesis has demonstrated that there is no consensus on how bleeding is assessed and graded in this patient group. Also it showed that the absolute immature platelet number may be a better alternative to the total platelet count to guide administration of platelet transfusions. Female sex, a previous history of a fungal infection, a high C-reactive protein, a high white cell count, a low platelet count, anaemia, impaired renal function, and recent clinically significant haemorrhage were all found to be independent risk factors for haemorrhage. Patients who were in complete remission from their haematological malignancy had a much lower risk of bleeding.

Introduction: Haematological malignancies (HMs) are a rare and diverse group of malignancies accounting for 9% of cancers globally. These group of malignancies differ by age, sex, subtypes, morphology and geography. The burden and the patterns of diversity of HMs is poorly understood in low and middle-income countries including Kenya. Aim: To analyse the time trends of incidence of haematological malignancies in Kenya by broad subtypes from 2000–2013 and to compare differences in trends of HMs between Nairobi and Uasin Gishu counties for the period 2007-2013. Methods: A retrospective study including all HMs for all ages and sex diagnosed in the period of 2000-2013. Information used was from two population based cancer registries; Eldoret and Nairobi cancer registry. Crude incidence rates were directly standardized with the world population to obtain the age-standardized rates (ASR). Sex rate ratios (SRR) and incident rate ratios (IRR) were then calculated to compare the number of excess cases between sexes and counties. Ms Excel and STATA13 software were used to conduct a time trend analysis of haematological malignancies by broad subtypes of non-Hodgkin lymphoma (NHL), Hodgkin lymphoma (HL), myeloma and leukaemia. Using the estimated annual percentage change (APC), increase or decrease in trends of HMs was determined. Results: In Kenya, the mean age at diagnosis for all HMs was 32 years. NHL was the most commonly diagnosed HM in Kenya accounting for 43.6% of the cases. The main basis of diagnosis for NHL and HL cases was by cytology while for myeloma and leukaemia was by histology. A male excess was noted in the NHL, myeloma and leukaemia cases with an exemption of a female excess in the HL cases. Trends in incidence of HMs in Kenya increased by 9.8% with the myeloma subtype contributing greatly to the observed increase. By counties, Uasin Gishu county reported a higher number of HM cases per 100000 than Nairobi county (Uasin Gishu-97.6 per 100000 and Nairobi-69.9 per 100000). On the contrary, Nairobi marked a higher increase in trends of HMs than Uasin Gishu county. Conclusion: Trends of haematological malignancies are increasing in Kenya and special attention needs to be given to these under-reported group of malignancies. Finally, this study does support the dire need for a national cancer registry in the country.

L'interleukine-6 (IL-6) exerce des propriétés biologiques multiples telles que l'activation des cellules immunocompétentes, l'activation de la réponse inflammatoire et l'hématopoïèse. Produite également par les cellules tumorales, l'IL-6 impacte la prolifération, la différenciation et la survie de ces dernières. L'IL-6 représente donc depuis plusieurs années une cible thérapeutique pertinente. Dans la première partie de ce travail, nous avons exploré une nouvelle piste potentielle pour bloquer l'activité biologique de l'IL-6, en utilisant un antagoniste naturel que notre équipe a identifié dans plusieurs lignées d'adénocarcinomes du rein, à savoir la molécule tronquée tIL-6. Suite à l'évaluation comparée de deux systèmes d'expression (E. coli versus CHO), nous avons retenu les cellules CHO comme source de production de fractions enrichies en tIL-6 par chromatographie de gel d'exclusion. Disposant d'un panel d'adénocarcinomes de rein (ACHN, Caki1, CLB CHA, CLB VER) et d'une lignée érythroleucémique (TF1), l'activité fonctionnelle de tIL-6 in vitro a été étudiée sur (1) la signalisation IL-6 induite, (2) la prolifération cellulaire IL-6 induite, la survie cellulaire et (4) la modulation de l'expression de protéines relevantes de l'apoptose. La molécule tIL-6 bloque la phosphorylation de la tyrosine Tyr705 de STAT3, qui est un des éléments clés de la voie de signalisation de l'IL-6. Nous rapportons également une autre observation nouvelle indiquant que tIL-6 exerce un effet pro-apoptotique sur certaines lignées RCC. Dans la seconde partie de notre étude, l'impact d'un Ac Mo anti IL-6 dans la réversion de la résistance aux cytotoxiques ou à la radiothérapie a été étudié. Nos résultats démontrent que la voie IL-6 ne constituerait pas un mécanisme majeur de résistance / Interleukin-6 (IL-6) plays numerous physiological roles including haematopoiesis, immune response and inflammation, but also plays a role in modulating cell growth, differentiation and survival of tumors cells. The first goal of the present study was to investigate on the potential role of the truncated protein IL-6 (tIL-6) encoded by the spliced IL-6 mRNA discovered in renal carcinoma cells (RCC). The R&D program was designed based on an industrial approach, aiming at reaching the decision stage to enter or not into preclinical development. Firstly two different expression systems were investigated (E. coli versus CHO cell line). The mammalian expression system was selected as the protein source since a recombinant glycosylated tIL-6 with a molecular weight similar to the predicted natural molecule was obtained from enriched fractions following size exclusion chromatography. Secondly by using a cell line panel including renal carcinoma cells (ACHN, Caki-1, CLB CHA, CLB-VER ) and an erythroleucemic cell line (TF1), in vitro tIL-6 functional activity were analyzed on (1) IL-6 induced signaling, (2) IL-6 induced cell proliferation, (3) on cell survival and also (4) on expression of specific set of proteins involved in apoptosis pathways. The truncated IL-6 was found inhibit IL-6 induced STAT3 Tyr705 and to induce apoptosis in some RCC cell lines which could be depending on IL-6 expression. Understanding more precisely the role of natural truncated IL-6 and its impact in cell tumour growth control will be a major issue in the development of innovative approach to antagonize directly or not IL6. The second goal of the present study was to investigate on reversing resistance of cancer cell lines to cytotoxics or ionizing radiations through the use of a monoclonal antibody directed against IL-6. Our data support the fact that IL-6 is not the preponderant actor of cell resistance to cytotoxics and ionizing radiations, which seems to be regulated by a complex network of proteins