Lesion detection and classification in breast cancer: evaluation of approaches based on morphological features, tracer kinetic modelling and semi-quantitative parameters in MR functional imaging (DCE-MRI)

The diagnosis, grading and classification of tumours has benefited considerably from the development of DCE-MRI which is now essential to the adequate clinical management of many tumour types due to its capability in detecting active angiogenesis. Several strategies have been proposed for DCE-MRI evaluation. Visual inspection of contrast agent concentration curves vs time is a very simple yet operator dependent procedure, therefore more objective approaches have been developed in order to facilitate comparison between studies. In so called model free approaches, descriptive or heuristic information extracted from time series raw data have been used for tissue classification. The main issue concerning these schemes is that they have not a direct interpretation in terms of physiological properties of the tissues. On the other hand, model based investigations typically involve compartmental tracer kinetic modelling and pixel-by-pixel estimation of kinetic parameters via non-linear regression applied on region of interests opportunely selected by the physician. This approach has the advantage to provide parameters directly related to the pathophysiological properties of the tissue such as vessel permeability, local regional blood flow, extraction fraction, concentration gradient between plasma and extravascular-extracellular space. Anyway, nonlinear modelling is computational demanding and the accuracy of the estimates can be affected by the signal-to-noise ratio and by the initial solutions. The principal aim of this thesis is investigate the use of semi-quantitative and quantitative parameters for segmentation and classification of breast lesion. The objectives can be subdivided as follow: describe the principal techniques to evaluate time intensity curve in DCE-MRI with focus on kinetic model proposed in literature; to evaluate the influence in parametrization choice for a classic bi-compartmental kinetic models; to evaluate the performance of a method for simultaneous tracer kinetic modelling and pixel classification; to evaluate performance of machine learning techniques training for segmentation and classification of breast lesion.

Trasmissione post-natale del citomegalovirus attraverso il latte materno al neonato VLBW

Introduction Postnatal human cytomegalovirus (CMV) infection is usually asymptomatic in term babies, while preterm infants are more susceptible to symptomatic CMV infection. Breastfeeding plays a dominant role in the epidemiology of transmission of postnatal CMV infection, but the risk factors of symptomatic CMV infection in preterm infants are unknown. Patients and Methods Between December 2003 and August 2006, eighty Very Low Birth Weight (VLBW) preterm infants (gestational age ≤ 32 weeks and birth weight < 1500 g), admitted to the Neonatal Intensive Care Unit of St Orsola-Malpighi General Hospital, Bologna were recruited. All of them were breastfed for at least one month. During the first week of life, serological test for CMV was performed on maternal blood. Furthermore, urinary CMV culture was performed in all the infants in order to exclude a congenital CMV infection. Urine samples from each infant were collected and processed for CMV culture once a week. Once every 15 days a blood sample was taken from each infant to evaluate the complete blood count, the hepatic function and the C reactive protein. In addition, samples of fresh breast milk were processed weekly for CMV culture. A genetic analysis of virus variant was performed in the urine of the infected infants and in their mother’s milk to confirm the origin of infection. Results We evaluated 80 VLBW infants and their 68 mothers. Fifty-three mothers (78%) were positive for CMV IgG antibodies, and 15 (22%) were seronegative. In the seronegative group, CMV was never isolated in breast milk, and none of the 18 infants developed viruria; in the seropositive group, CMV was isolated in 21 out of 53 (40%) mother’s milk. CMV was detected in the urine samples of 9 out of 26 (35%) preterm infants, who were born from 21 virolactia positive mothers. Six of these infants had clinically asymptomatic CMV infection, while 3 showed a sepsis-like illness with bradycardia, tachypnea and repeated desaturations. Eight out of nine infants showed abnormal hematologic values. The detection of neutropenia was strictly related to CMV infection (8/9 infected infants vs 17/53 non infected infants, P<.005), such as the detection of an increase in conjugated bilirubin (3/9 infected infants vs 2/53 non infected infants, P<.05). The degree of neutropenia was not different between the two groups (infected/non infected). The use of hemoderivatives (plasma and/or IgM–enriched immunoglobulin) in order to treat a suspected/certain infection in newborn with GE< 28 ws was seen as protective against CMV infection (1/4 infected infants vs 18/20 non infected infants [GE<28 ws]; P<.05). Furthermore, bronchopulmonary dysplasia (defined both as oxygen-dependency at 30 days of life and 36 ws of postmenstrual age) correlated with symptomatic infection (3/3 symptomatic vs 0/6 asymptomatic: P<.05). Conclusion Our data suggest that CMV infection transmitted to preterm newborn through human milk is always asymptomatic when newborns are clinically stable. Otherwise, the infection can worsen a preexisting disease such as bronchopulmonary dysplasia. Human milk offers many nutritional and psychological advantages to preterm newborns: according to our data, there’s no reason to contraindicate it neither to pasteurize the milk of all the mothers of preterm infants who are CMV seropositive.

The role of interleuchin 6 (IL-6) in the promotion of an aggressive and stem cell-like phenotype in human breast cancer cells and in stem/progenitor cells expanded in vitro as mammospheres

High serum levels of Interleukin-6 (IL-6) correlate with poor outcome in breast cancer patients. However no data are available on the relationship between IL-6 and stem/progenitor cells which may fuel the genesis of breast cancer in vivo. Herein, we address this issue in mammospheres (MS), multi-cellular structures enriched in stem/progenitor cells of the mammary gland, and also in MCF-7 breast cancer cells. We show that MS from node invasive breast carcinoma tissues express IL-6 mRNA at higher levels than MS from matched non-neoplastic mammary glands. We find that IL-6 mRNA is detectable only in basal-like breast carcinoma tissues, an aggressive variant showing stem cell features. Our results reveal that IL-6 triggers a Notch-3-dependent up-regulation of the Notch ligand Jagged-1, whose interaction with Notch-3 promotes the growth of MS and MCF-7 derived spheroids. Moreover, IL-6 induces a Notch-3-dependent up-regulation of the carbonic anhydrase IX gene, which promotes a hypoxia-resistant/invasive phenotype in MCF-7 cells and MS. Finally, an autocrine IL-6 loop relies upon Notch-3 activity to sustain the aggressive features of MCF-7-derived hypoxia-selected cells. In conclusion, our data support the hypothesis that IL-6 induces malignant features in Notch-3 expressing, stem/progenitor cells from human ductal breast carcinoma and normal mammary gland.

Stem cell pathways in the basal-like breast carcinoma: the role of beta-catenin and HIF-1alpha

Basal-like tumor is an aggressive breast carcinoma subtype that displays an expression signature similar to that of the basal/myoepithelial cells of the breast tissue. Basal-like carcinoma are characterized by over-expression of the Epidermal Growth Factor receptor (EGFR), high frequency of p53 mutations, cytoplasmic/nuclear localization of beta-catenin, overexpression of the Hypoxia inducible factor (HIF)-1alpha target Carbonic Anhydrase isoenzime 9 (CA9) and a gene expression pattern similar to that of normal and cancer stem cells, including the over-expression of the mammary stem cell markers CD44. In this study we investigated the role of p53, EGFR, beta-catenin and HIF-1alpha in the regulation of stem cell features and genes associated with the basal-like gene expression profile. The findings reported in this investigation indicate that p53 inactivation in ductal breast carcinoma cells leads to increased EGFR mRNA and protein levels. In our experimental model, EGFR overexpression induces beta-catenin cytoplasmatic stabilization and transcriptional activity and, by that, leads to increased aggressive features including mammosphere (MS) forming and growth capacity, invasive potential and overexpression of the mammary stem cell gene CD44. Moreover we found that EGFR/beta-catenin axis promotes hypoxia survival in breast carcinoma cells via increased CA9 expression. Indeed beta-catenin positively regulates CA9 expression upon hypoxia exposure. Interestingly we found that beta-catenin inhibits HIF-1alpha transcriptional activity. Looking for the mechanism, we found that CA9 expression is promoted by HIF-1alpha and cytoplasmatic beta-catenin further increased it post-transcriptionally, via direct mRNA binding and stabilization. These data reveal a functional beta-catenin/HIF-1alpha interplay among hallmarks of basal-like tumors and unveil a new functional role for cytoplasmic beta-catenin in the phenotype of such tumors. Therefore it can be proposed that the interplay here described among EGFR/beta-catenin and HIF-1alpha may play a role in breast cancer stem cell survival and function.

Perfluoroalkylated substances in food matrices: development of mass spectrometry based analytical methods and preliminary monitoring

Perfluoroalkylated substances are a group of chemicals that have been largely employed during the last 60 years in several applications, widely spreading and accumulating in the environment due to their extreme resistance to degradation. As a consequence, they have been found also in various types of food as well as in drinking water, proving that they can easily reach humans through the diet. The available information concerning their adverse effects on health has recently increased the interest towards these contaminants and highlighted the importance of investigating all the potential sources of human exposure, among which diet was proved to be the most relevant. This need has been underlined by the European Union through Recommendation 2010/161/EU: in this document, Member States were called to monitor their presence of in food, producing accurate estimations of human exposure. The purpose of the research presented in this thesis, which is the result of a partnership between an Italian and a French laboratory, was to develop reliable tools for the analysis of these pollutants in food, to be used for generating data on potentially contaminated matrices. An efficient method based on liquid chromatography-mass spectrometry for the detection of 16 different perfluorinated compounds in milk has been validated in accordance with current European regulation guidelines (2002/657/EC) and is currently under evaluation for ISO 17025 accreditation. The proposed technique was applied to cow, powder and human breast milk samples from Italy and France to produce a preliminary monitoring on the presence of these contaminants. In accordance with the above mentioned European Recommendation, this project led also to the development of a promising technique for the quantification of some precursors of these substances in fish. This method showed extremely satisfying performances in terms of linearity and limits of detection, and will be useful for future surveys.

Evaluation of Antitumoral activity of bone targeted drugs/conventional chemotherapies and identification of biomarkers for the selection of patients with breast cancer for the bone targeted therapy in adjuvant setting

Bone metastases are responsible for different clinical complications defined as skeletal-related events (SREs) such as pathologic fractures, spinal cord compression, hypercalcaemia, bone marrow infiltration and severe bone pain requiring palliative radiotherapy. The general aim of these three years research period was to improve the management of patients with bone metastases through two different approaches of translational research. Firstly in vitro preclinical tests were conducted on breast cancer cells and on indirect co-colture of cancer cells and osteoclasts to evaluate bone targeted therapy singly and in combination with conventional chemotherapy. The study suggests that zoledronic acid has an antitumor activity in breast cancer cell lines. Its mechanism of action involves the decrease of RAS and RHO, as in osteoclasts. Repeated treatment enhances antitumor activity compared to non-repeated treatment. Furthermore the combination Zoledronic Acid + Cisplatin induced a high antitumoral activity in the two triple-negative lines MDA-MB-231 and BRC-230. The p21, pMAPK and m-TOR pathways were regulated by this combined treatment, particularly at lower Cisplatin doses. A co-colture system to test the activity of bone-targeted molecules on monocytes-breast conditioned by breast cancer cells was also developed. Another important criticism of the treatment of breast cancer patients, is the selection of patients who will benefit of bone targeted therapy in the adjuvant setting. A retrospective case-control study on breast cancer patients to find new predictive markers of bone metastases in the primary tumors was performed. Eight markers were evaluated and TFF1 and CXCR4 were found to discriminate between patients with relapse to bone respect to patients with no evidence of disease. In particular TFF1 was the most accurate marker reaching a sensitivity of 63% and a specificity of 79%. This marker could be a useful tool for clinicians to select patients who could benefit for bone targeted therapy in adjuvant setting.

Valutazione della presenza di contaminanti perfluoroalchilici in alimenti destinati al consumo umano

Le sostanze perfluoralchiliche (PFAS), composti fluorurati ampiamente utilizzati negli ultimi anni in diverse applicazioni industriali e commerciali, sono ritrovati diffusamente nell’ambiente e in diverse specie animali. Recentemente i PFAS hanno destato preoccupazione anche per la salute umana. Il rischio di esposizione è principalmente legato alla dieta (i prodotti ittici sembrano essere gli alimenti più contaminati). Lo scopo di questo lavoro è stato quello di valutare la presenza del perfluorottanosulfonato (PFOS) e dell’acido perfluorottanoico (PFOA), in diverse matrici alimentari: latte vaccino commercialmente disponibile in Italia, latte materno italiano, diverse specie di pesce commercialmente disponibili in Italia e 140 branzini di diverse aree (principalmente Mediterraneo). I campioni di latte sono stati trattati con estrazione liquido-liquido seguita da due fasi di purificazione mediante cartucce SPE prima dell’iniezione nell’UPLC-MS/MS. L’analisi del latte vaccino ha evidenziato una contaminazione diffusa di PFOS, ma a basse concentrazioni (fino a 97 ng/L), mentre il PFOA è stato ritrovato raramente. In questo studio, in grado di individuare anche i livelli delle ultra-tracce, sono state osservate nel latte materno concentrazioni di 15-288 ng/L per il PFOS e di 24-241 ng/LPFOA. Le concentrazioni e le frequenze più alte, per entrambi i PFAS, sono stati ritrovate in campioni di latte forniti da donne primipare, suggerendo un rischio di esposizione per i primogeniti. Il metodo utilizzato per i campioni di pesce era basato su un’estrazione con solvente organico seguita da due fasi di purificazione: una con i sali e una con fase solida dispersiva. L’estratto, analizzato in UPLC-MS/MS, ha confermato la contaminazione di questa matrice a livelli significativi, ma anche l’alta variabilità delle concentrazioni misurate. Il monitoraggio monospecie ha mostrato una contaminazione rilevante (PFOS 11,1- > 10000 ng/L; PFOA < 9-487 ng/L), soprattutto nei branzini pescati, rispetto a quelli allevati.