Prevention of central nervous system (CNS) relapse is critical for cure of childhood B-cell precursor acute lymphoblastic leukaemia (BCP-ALL). Despite this, mechanisms of CNS infiltration are poorly understood and the timing, frequency and properties of BCP-ALL blasts entering the CNS compartment are unknown. We investigated the CNS-engrafting potential of BCP-ALL cells xenotransplanted into immunodeficient NOD.Cg-Prkdc^scidIl2rg^tm1Wjl/SzJ mice. CNS engraftment was seen in 23/29 diagnostic samples (79%), 2/2 from patients with overt CNS disease and 21/27 (78%) from patients thought to be CNS-negative by diagnostic lumbar puncture. Histological findings mimic human pathology and demonstrate that leukaemic cells primarily transit the blood-cerebrospinal-fluid barrier sitting in close proximity to the dural sinuses – the site of recently discovered CNS lymphatics. Retrieval of blasts from the CNS showed no evidence for chemokine receptor-mediated selective trafficking. The high frequency of infiltration and lack of selective trafficking led us to postulate that CNS tropism is a generic property of leukaemic cells. To test this we performed serial dilution experiments, CNS engraftment was seen in 5/6 mice following transplantation of as few as 10 leukaemic cells. Finally, clonal tracking techniques confirmed the polyclonal nature of CNS infiltrating cells with multiple clones engrafting in both the CNS and periphery. Overall, these findings suggest that sub-clinical seeding of the CNS is likely to be present in the majority of BCP-ALL patients at original diagnosis and efforts to prevent CNS relapse should concentrate on augmenting effective eradication of disease from this site, rather than targeting entry mechanisms.

Hypoxia-inducible factor-1α (HIF-1α) has been shown to be involved in cancer metastasis in several cancer types. There is however conflicting evidence of HIF-1α expression with oral cancer prognosis. Therefore, this study set out to investigate HIF-1α overexpression and its relationship with the aggressiveness and grade of oral squamous cell carcinoma (OSCC) and to explore the diagnostic potential of HIF-1α overexpression in OSCC in a cohort of Pakistani patients. Immunostaining of HIF-1α was performed on 54 OSCC and 14 normal oral mucosa (NOM) tissue samples and various cut-offs were used to evaluate its immunohistochemical expression. HIF-1α expression in OSCC samples was significantly higher than in controls, with minimal immunoreactivity in NOM. HIF-1α overexpression was significantly associated with increased tumor size (p = 0.046). However, no association was found between HIF-1α overexpression and increasing Broder’s histological grade or TNM stage. The cut-off >10% cells with moderate to marked intensity carried a sensitivity of 70% and a specificity of 100% to distinguish between tumor and control. ROC curve analysis of HIF-1α weighted histoscores showedHIF-1α overexpression as a highly sensitive and specific diagnostic test (p < 0.001, AUC = 0.833). HIF-1α overexpression is a tumor-specific finding associated with increased tumor size and carries a potential diagnostic role.

Haemoglobin (Hb) electrophoresis is a method of blood testing used to detect thalassaemia. However, the interpretation of the result of the electrophoresis test itself is a complex task. Expert haematologists, specifically in developing countries, are relatively few in number and are usually overburdened. To assist them with their workload, in this paper we present a novel method for the automated assessment of thalassaemia using Hb electrophoresis images. Moreover, in this study we compile a large Hb electrophoresis image dataset, consisting of 103 strips containing 524 electrophoresis images with a clear consensus on the quality of electrophoresis obtained from 824 subjects. The proposed methodology is split into two parts: (1) single-patient electrophoresis image segmentation by means of the lane extraction technique, and (2) binary classification (normal or abnormal) of the electrophoresis images using state-of-the-art deep convolutional neural networks (CNNs) and using the concept of transfer learning. Image processing techniques including filtering and morphological operations are applied for object detection and lane extraction to automatically separate the lanes and classify them using CNN models. Seven different CNN models (ResNet18, ResNet50, ResNet101, InceptionV3, DenseNet201, SqueezeNet and MobileNetV2) were investigated in this study. InceptionV3 outperformed the other CNNs in detecting thalassaemia using Hb electrophoresis images. The accuracy, precision, recall, f1-score, and specificity in the detection of thalassaemia obtained with the InceptionV3 model were 95.8%, 95.84%, 95.8%, 95.8% and 95.8%, respectively. MobileNetV2 demonstrated an accuracy, precision, recall, f1-score, and specificity of 95.72%, 95.73%, 95.72%, 95.7% and 95.72% respectively. Its performance was comparable with the best performing model, InceptionV3. Since it is a very shallow network, MobileNetV2 also provides the least latency in processing a single-patient image and it can be suitably used for mobile applications. The proposed approach, which has shown very high classification accuracy, will assist in the rapid and robust detection of thalassaemia using Hb electrophoresis images.