Endometrial cancer is the most common gynecologic cancer in developed countries and the world's sixth most common cancer in women. In almost 80% of cases, endometrial cancer is of type I (endometrioid histological type), mostly of low grade, while the remaining 20% of the tumors is called type II (non-endometrioid) and normally has poor prognosis. Currently, grading classification of type I endometrial cancer takes place through pathologic analysis of tumor morphology, but often the assessment of the type and tumor grading is not sufficient to predict prognosis. In order to develop new and more effective prognostic tools the identification of genetic and/or molecular variables associated to the phenotype and profiles of aggressiveness of these tumors is necessary. The purpose of this study is to develop a prediction method based on genes mutational status, useful to stratify the endometrial tumors in the categories "good prognosis" and "poor prognosis". The analysis was performed on 89 samples of endometrial cancer of different degrees. High-throughput sequencing was performed on Illumina MiSeq sequencer using the "Trusight Tumor 26 kit", which allows the analysis of somatic mutations described to 174 amplicons within 26 genes selected among those most associated with solid tumors in literature. An unsupervised clustering analysis was carried out to divide samples into two groups, based on the mutational profile. Survival analysis performed showed a basically different behavior of the two groups (18% deaths in cluster 1 and 11% in cluster 2, 23% recurrence in cluster 1 and 11% in cluster 2). The analysis also showed the association of the clustering with the grading of the tumors, in particular distinguishing perfectly grade G1 samples and placing them all in the “good prognosis” cluster 1. The analysis of frequency of the mutated genes and their number of mutations in clusterized samples showed that APC, CTNNB1, KRAS, PIK3CA, PTEN mutation status was different in the two groups. Mutations on CTNNB1, KRAS, PIK3CA, SMAD4 and TP53 were also found associated with the different grading of endometrial tumors. The study therefore proposes a new approach based on molecular profiling, that trough the sequencing of a small panel of genes, could support the histological analysis in endometrial cancer outcome prediction, especially in doubtful cases.
Il cancro dell’endometrio è il tumore ginecologico più comune nei paesi sviluppati ed è al mondo il sesto tumore più comune nelle donne. In quasi l'80% dei casi, il tumore dell’endometrio è di tipo I (istotipo endometrioide), per lo più di basso grado, mentre il restante 20% dei tumori si definisce di tipo II (non endometrioide) e normalmente ha prognosi più aggressiva rispetto ai tumori di tipo I. Attualmente, la definizione del grado del tumore di tipo I avviene tramite analisi anatomo-patologica della morfologia tumorale, ma spesso la valutazione del tipo e del grado del tumore non è sufficiente a predire la prognosi. L’identificazione di variabili genetiche e/o molecolari associate al fenotipo e ai profili di aggressività di questi tumori è necessaria al fine di sviluppare nuovi e più efficaci strumenti prognostici. Lo scopo di questo studio è lo sviluppo di un metodo di predizione basato sull'analisi mutazionale, che sia utile per stratificare i tumori dell'endometrio nelle categorie “buona prognosi" e “cattiva prognosi”. L'analisi è stata effettuata su 89 campioni di tumore dell'endometrio a diversi gradi. Ha previsto il sequenziamento high-throughput, mediante sequenziatore MiSeq Illumina, utilizzando il kit “Trusight Tumor” il quale consente l'analisi delle mutazioni somatiche descritte su 174 ampliconi all’interno di 26 geni selezionati tra quelli maggiormente associati a tumori solidi in letteratura. Un’analisi di clustering non supervisionato è stata effettuata per suddividere i campioni in due gruppi, sulla base del profilo mutazionale. L'analisi di sopravvivenza effettuata sui 2 gruppi ha mostrato un comportamento tendenzialmente differente dei due gruppi (Morti 18% nel cluster 1 e 11% nel cluster 2, Recidive 23 % nel cluster 1 e 11% nel cluster 2). L’analisi ha inoltre mostrato l'associazione della clusterizzazione su base molecolare con il grado del tumore, in particolare distinguendo perfettamente i campioni di grado G1 e collocandoli tutti nel cluster 1 a buona prognosi. L’analisi di frequenza dei geni mutati e del numero di mutazioni di ciascuno nei campioni dei due clusters ha evidenziato come i geni APC, CTNNB1, KRAS, PIK3CA, PTEN siano differentemente mutati nei due gruppi Le mutazioni sui geni CTNNB1, KRAS, PIK3CA, SMAD4 e TP53 sono risultate inoltre associate al diverso grado dei tumori. Lo studio propone quindi un nuovo approccio su base molecolare, che prevedendo il sequenziamento di un piccolo pannello di geni, potrebbe affiancare l'analisi istologica del tumore dell'endometrio nella predizione dell'outcome tumorale, soprattutto dei casi a prognosi più incerta.
Un nuovo approccio molecolare per la prognosi del tumore dell'endometrio
TORRICELLI, FEDERICA
2017-07-19
Abstract
Endometrial cancer is the most common gynecologic cancer in developed countries and the world's sixth most common cancer in women. In almost 80% of cases, endometrial cancer is of type I (endometrioid histological type), mostly of low grade, while the remaining 20% of the tumors is called type II (non-endometrioid) and normally has poor prognosis. Currently, grading classification of type I endometrial cancer takes place through pathologic analysis of tumor morphology, but often the assessment of the type and tumor grading is not sufficient to predict prognosis. In order to develop new and more effective prognostic tools the identification of genetic and/or molecular variables associated to the phenotype and profiles of aggressiveness of these tumors is necessary. The purpose of this study is to develop a prediction method based on genes mutational status, useful to stratify the endometrial tumors in the categories "good prognosis" and "poor prognosis". The analysis was performed on 89 samples of endometrial cancer of different degrees. High-throughput sequencing was performed on Illumina MiSeq sequencer using the "Trusight Tumor 26 kit", which allows the analysis of somatic mutations described to 174 amplicons within 26 genes selected among those most associated with solid tumors in literature. An unsupervised clustering analysis was carried out to divide samples into two groups, based on the mutational profile. Survival analysis performed showed a basically different behavior of the two groups (18% deaths in cluster 1 and 11% in cluster 2, 23% recurrence in cluster 1 and 11% in cluster 2). The analysis also showed the association of the clustering with the grading of the tumors, in particular distinguishing perfectly grade G1 samples and placing them all in the “good prognosis” cluster 1. The analysis of frequency of the mutated genes and their number of mutations in clusterized samples showed that APC, CTNNB1, KRAS, PIK3CA, PTEN mutation status was different in the two groups. Mutations on CTNNB1, KRAS, PIK3CA, SMAD4 and TP53 were also found associated with the different grading of endometrial tumors. The study therefore proposes a new approach based on molecular profiling, that trough the sequencing of a small panel of genes, could support the histological analysis in endometrial cancer outcome prediction, especially in doubtful cases.File | Dimensione | Formato | |
---|---|---|---|
Torricelli_Tesi dottorato_definitiva_con_frontespizio.pdf
accesso aperto
Descrizione: tesi di dottorato
Dimensione
2.26 MB
Formato
Adobe PDF
|
2.26 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.