Carcinomas have a distinct structure which mimics that of normal tissues and comprises two distinct but interdependent compartments: the parenchyma (neoplastic cells) and the stroma. Stroma is interposed between malignant cells and normal host tissues and is essential for tumor growth. Stroma is largely a product of the host and its development is induced by tumor cell-host interactions. Thus, it comprises nonmalignant supporting tissue and includes leaked plasma and plasma proteins, interstitial collagens and fibrin. Moreover it includes three types of cells: endothelial cells forming new blood vessels, fibroblasts that reside also in normal connective tissue, and inflammatory cells that are derived from the blood. Tumors markedly differ from each other in quantitative stromal content, with significant variations in stromal composition from one area to another even within a single tumor. In this thesis different carcinoma models were analyzed in vivo - by contrast enhanced magnetic resonance imaging (MRI), and by (18)F-fluorodeoxy-glucose (FDG) positron emission tomography (PET) - focusing on the role of the stromal compartment. Since contrast enhanced MRI and FDG-PET are sensitive to tumor angiogenesis and glucose metabolism respectively, our studies investigated the correlation between perfusion and glucose metabolism. The same tumor models were also evaluated ex vivo - by histology and immunohistochemistry techniques. It is well known that cancer cells are characterized by anaerobic metabolism with high glucose consumption, both in hypoxic condition and even in high oxygen tension (Warburg effect). On the other hand, it has been recently reported that in carcinomas a complementary metabolism does exist between the epithelial neoplastic cells and the stromal fibroblasts. In fact, stromal fibroblasts have shown aerobic metabolism, consistent with higher oxygen availability due to the associated vascular supply, and the capability of buffering and recycling products of anaerobic metabolism to sustain cancer cell survival. In the present thesis, the complementary stromal/epithelial metabolism was confirmed by the resulting different FDG-uptake in two carcinoma models characterized by a markedly different stromal content1. These findings were also supported by immunohistochemical examination with markers specific for neoangiogenesis and glucose transporters. On the same experimental models, a corresponding different perfusion was observed by contrast enhanced MRI2. The different distribution of the contrast agents proved to be related to stromal content, which presumably produced also a different washout pattern of the contrast agent itself. Consistently, in an additional study on experimental mammary tumors, we demonstrated that the washout of the contrast agents mainly occurs by the venous system and not by lymphatic vessels3. Moreover, three different breast tumor models (a spontaneous and an implanted carcinoma, and a mesenchymal tumor), characterized by different vascularization, were evaluated. In these models perfusion and metabolism resulted to be complementary: tumors (and tumor areas) that were characterized by higher MR contrast enhancement showed lower FDG-uptake and vice-versa4. Such effect could constitute a potential risk of tumor volume underestimation when FDGPET is used as a single imaging modality to assess tumor boundaries and to delineate the target volume in radiotherapy planning5. Finally, tumor stroma evolution was assessed during antiangiogenic therapy6. Contrast enhanced MRI and histology revealed that prolonged treatment could promote an abnormal stromal development at the periphery of carcinomas, suggesting that cancer-associated stroma can have a role in the adaptive response to treatment. In conclusion, the reported studies on experimental models showed that the compartmental architecture of carcinomas, i.e. the neoplastic cell and the cancer-associated stroma, affects the sensitivity of two major cancer imaging methods, contrast enhanced MRI and FDG-PET. Contrast enhanced MRI appeared more sensitive to the presence of cancer-associated stroma due to perfusion. FDG-PET appeared more sensitive to the presence of cancer cells, due to glucose metabolism. This “complementary sensitivity” suggests a combined application of the two modalities for a comprehensive evaluation of carcinomas. Furthermore, the contrast enhanced MRI modality allowed to evaluate both washout mechanisms and the stromal modifications occurring during antiangiogenic therapy. The reported findings, if confirmed in clinical studies, could be applied in diagnosis, treatment planning and therapy assessment of human carcinomas. In any case they are relevant in experimental studies considering the emerging role of cancer-associated stroma to understand cancer progression and eventually to develop new targeted therapies.

Imaging multimodale in carcinomi sperimentali: il ruolo della componente stromale

FARACE, Paolo
2009

Abstract

Carcinomas have a distinct structure which mimics that of normal tissues and comprises two distinct but interdependent compartments: the parenchyma (neoplastic cells) and the stroma. Stroma is interposed between malignant cells and normal host tissues and is essential for tumor growth. Stroma is largely a product of the host and its development is induced by tumor cell-host interactions. Thus, it comprises nonmalignant supporting tissue and includes leaked plasma and plasma proteins, interstitial collagens and fibrin. Moreover it includes three types of cells: endothelial cells forming new blood vessels, fibroblasts that reside also in normal connective tissue, and inflammatory cells that are derived from the blood. Tumors markedly differ from each other in quantitative stromal content, with significant variations in stromal composition from one area to another even within a single tumor. In this thesis different carcinoma models were analyzed in vivo - by contrast enhanced magnetic resonance imaging (MRI), and by (18)F-fluorodeoxy-glucose (FDG) positron emission tomography (PET) - focusing on the role of the stromal compartment. Since contrast enhanced MRI and FDG-PET are sensitive to tumor angiogenesis and glucose metabolism respectively, our studies investigated the correlation between perfusion and glucose metabolism. The same tumor models were also evaluated ex vivo - by histology and immunohistochemistry techniques. It is well known that cancer cells are characterized by anaerobic metabolism with high glucose consumption, both in hypoxic condition and even in high oxygen tension (Warburg effect). On the other hand, it has been recently reported that in carcinomas a complementary metabolism does exist between the epithelial neoplastic cells and the stromal fibroblasts. In fact, stromal fibroblasts have shown aerobic metabolism, consistent with higher oxygen availability due to the associated vascular supply, and the capability of buffering and recycling products of anaerobic metabolism to sustain cancer cell survival. In the present thesis, the complementary stromal/epithelial metabolism was confirmed by the resulting different FDG-uptake in two carcinoma models characterized by a markedly different stromal content1. These findings were also supported by immunohistochemical examination with markers specific for neoangiogenesis and glucose transporters. On the same experimental models, a corresponding different perfusion was observed by contrast enhanced MRI2. The different distribution of the contrast agents proved to be related to stromal content, which presumably produced also a different washout pattern of the contrast agent itself. Consistently, in an additional study on experimental mammary tumors, we demonstrated that the washout of the contrast agents mainly occurs by the venous system and not by lymphatic vessels3. Moreover, three different breast tumor models (a spontaneous and an implanted carcinoma, and a mesenchymal tumor), characterized by different vascularization, were evaluated. In these models perfusion and metabolism resulted to be complementary: tumors (and tumor areas) that were characterized by higher MR contrast enhancement showed lower FDG-uptake and vice-versa4. Such effect could constitute a potential risk of tumor volume underestimation when FDGPET is used as a single imaging modality to assess tumor boundaries and to delineate the target volume in radiotherapy planning5. Finally, tumor stroma evolution was assessed during antiangiogenic therapy6. Contrast enhanced MRI and histology revealed that prolonged treatment could promote an abnormal stromal development at the periphery of carcinomas, suggesting that cancer-associated stroma can have a role in the adaptive response to treatment. In conclusion, the reported studies on experimental models showed that the compartmental architecture of carcinomas, i.e. the neoplastic cell and the cancer-associated stroma, affects the sensitivity of two major cancer imaging methods, contrast enhanced MRI and FDG-PET. Contrast enhanced MRI appeared more sensitive to the presence of cancer-associated stroma due to perfusion. FDG-PET appeared more sensitive to the presence of cancer cells, due to glucose metabolism. This “complementary sensitivity” suggests a combined application of the two modalities for a comprehensive evaluation of carcinomas. Furthermore, the contrast enhanced MRI modality allowed to evaluate both washout mechanisms and the stromal modifications occurring during antiangiogenic therapy. The reported findings, if confirmed in clinical studies, could be applied in diagnosis, treatment planning and therapy assessment of human carcinomas. In any case they are relevant in experimental studies considering the emerging role of cancer-associated stroma to understand cancer progression and eventually to develop new targeted therapies.
2009
Italiano
imaging multimodale; carcinoma
64
File in questo prodotto:
File Dimensione Formato  
tesi Farace Paolo.pdf

accesso solo da BNCF e BNCR

Dimensione 6.4 MB
Formato Adobe PDF
6.4 MB Adobe PDF

I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.14242/113983
Il codice NBN di questa tesi è URN:NBN:IT:UNIVR-113983