Background: The local and nodal staging of rectal cancer is essential for the selection of the most appropriate surgical approach in these patients, especially after pCRT. The assessment of the number and dimensions of mesorectal lymph nodes has significant prognostic value, in spite of the limitations associated with the application of MRI in the assessment of lymp node involvement. Purpose: To compare the accuracy of different dimensional criteria for measuring the largest mesorectal lymph node in determining the presence of local and regional nodal metastasis after CRT in locally advanced rectal cancer, using histopathological analysis as the reference standard. Materials and methods: 139 patients with locally advanced rectal cancer, assessed by MRI before and after CRT (within one month), were included in the study. The obtained MRI images were independently assessed by two radiologists with no knowledge of the histopathological findings. Short-axis diameter, long-axis diameter and a third cranio-caudal axis of the largest mesorectal lymph node were measured. Additionally, the number of local and regional lymph nodes with a short-axis diameter ≥5 mm (ESGAR method) was assessed. The patients were subdivided into two groups, a derivation cohort (100 patients) and a validation cohort (39 patients). An agreement analysis was conducted for every type of measurement in the derivation cohort. Subsequently, the ROC curves for the short-axis diameter, the long-axis diameter and the volume of the largest lymph node were assessed to establish the best cut-off value through Youden’s index. The diagnostic performance of these methods and that of the ESGAR method were compared through McNemar’s test. Finally, the obtained results were tested on the validation cohort. Results: In the cohort of 100 patients, the agreement between the two readers was strong for three methods (κvolume=0.85; κlong-axis=0.80; κESGAR=0.88) and poor for the short-axis (κ=0.11). With a cut-off value of 5 mm for the short-axis and the long axis diameters, and of 95 mm³ for the volume of the largest lymph node, accuracy values of 66%, 69% and 73% were observed, respectively, whereas the ESGAR method was associated with an accuracy of 65%. The application of McNemar’s test showed no statistically significant differences. In the cohort of 39 patients, by applying the aforementioned cut-off values, the resulting accuracy was 66% for the short-axis, 77% for the long-axis, 74% for the volume and 67% for the number of lymph nodes (ESGAR method). Conclusion: The assessment of the volume or of the long-axis diameter of the largest visible lymph node after CRT in locally advanced rectal cancer is not inferior to the ESGAR method; therefore, it allows for a simpler and quicker evaluation of nodal status in these patients.
Introduzione: La stadiazione loco-regionale del cancro del retto è fondamentale per la scelta dell’approccio chirurgico più appropriato, in particolar modo dopo CRT neoadiuvante. La valutazione del numero e delle dimensioni dei linfonodi mesorettali ha un importante significato prognostico, nonostante la risonanza magnetica presenti in quest'ambito alcune notevoli limitazioni. Scopo dello studio: Confrontare l'accuratezza di diversi criteri di misurazione dimensionali del linfonodo mesorettale maggiore nel determinare la presenza di metastasi linfonodali loco-regionali dopo CRT nel tumore del retto localmente avanzato, usando l'analisi istopatologica come standard di riferimento. Materiali e metodi: 139 pazienti con cancro rettale localmente avanzato, valutati mediante risonanza magnetica prima e dopo CRT (entro un mese), sono stati inclusi nello studio. Le immagini di risonanza magnetica sono state valutate indipendentemente da due radiologi all'oscuro dei risultati dell’esame istopatologico. Sono stati misurati l'asse corto, l'asse lungo e un terzo asse cranio-caudale del linfonodo mesorettale più grande. Inoltre, è stato valutato il numero di linfonodi loco-regionali con asse corto ≥5 mm (metodo ESGAR). I pazienti sono stati suddivisi in una prima coorte di prova (100 pazienti) ed in una seconda coorte di validazione (39 pazienti). Nella prima coorte è stata effettuata un’analisi di concordanza per ogni tipo di misurazione. Successivamente sono state valutate le curve ROC per l’asse corto, l'asse lungo e il volume del linfonodo maggiore al fine di stabilire il miglior cut-off attraverso l’indice di Youden. L'accuratezza di questi metodi e del metodo ESGAR è stata comparata tramite il test di McNemar, considerando i risultati dell’esame istopatologico. I risultati ottenuti sono infine stati testati sulla coorte di validazione. Risultati: Nella coorte di 100 pazienti, la concordanza tra i due lettori è risultata elevata per tre metodi (κvolume=0.85; κasse lungo=0.80; κESGAR=0.88) e scarsa per l'asse corto (κ=0.11). Con un cut-off di 5 mm per l’asse corto e l’asse lungo, e di 95 mm³ per il volume del linfonodo maggiore, sono stati ottenuti valori di accuratezza rispettivamente del 66%, 69% e 73%, a fronte del 65% raggiunto dal metodo ESGAR. Applicando il test di McNemar non sono state riscontrate differenze statisticamente significative. Nella coorte di 39 pazienti, impiegando i suddetti cut-off, sono stati ottenuti valori di accuratezza per l’asse corto del 66%, per l’asse lungo del 77%, per il volume del 74% e per il numero di linfonodi (metodo ESGAR) del 67%. Conclusione: La valutazione del volume o del diametro dell’asse lungo del linfonodo maggiore dopo CRT nel tumore del retto localmente avanzato non è inferiore rispetto a quella prevista dal metodo ESGAR; pertanto, consente una più rapida e semplice valutazione dello stato linfonodale in tali pazienti.
Identificazione tramite RM delle metastasi linfonodali loco-regionali nel tumore del retto localmente avanzato dopo terapia neoadiuvante
HAJDUKOVIĆ, DEJAN
2021/2022
Abstract
Background: The local and nodal staging of rectal cancer is essential for the selection of the most appropriate surgical approach in these patients, especially after pCRT. The assessment of the number and dimensions of mesorectal lymph nodes has significant prognostic value, in spite of the limitations associated with the application of MRI in the assessment of lymp node involvement. Purpose: To compare the accuracy of different dimensional criteria for measuring the largest mesorectal lymph node in determining the presence of local and regional nodal metastasis after CRT in locally advanced rectal cancer, using histopathological analysis as the reference standard. Materials and methods: 139 patients with locally advanced rectal cancer, assessed by MRI before and after CRT (within one month), were included in the study. The obtained MRI images were independently assessed by two radiologists with no knowledge of the histopathological findings. Short-axis diameter, long-axis diameter and a third cranio-caudal axis of the largest mesorectal lymph node were measured. Additionally, the number of local and regional lymph nodes with a short-axis diameter ≥5 mm (ESGAR method) was assessed. The patients were subdivided into two groups, a derivation cohort (100 patients) and a validation cohort (39 patients). An agreement analysis was conducted for every type of measurement in the derivation cohort. Subsequently, the ROC curves for the short-axis diameter, the long-axis diameter and the volume of the largest lymph node were assessed to establish the best cut-off value through Youden’s index. The diagnostic performance of these methods and that of the ESGAR method were compared through McNemar’s test. Finally, the obtained results were tested on the validation cohort. Results: In the cohort of 100 patients, the agreement between the two readers was strong for three methods (κvolume=0.85; κlong-axis=0.80; κESGAR=0.88) and poor for the short-axis (κ=0.11). With a cut-off value of 5 mm for the short-axis and the long axis diameters, and of 95 mm³ for the volume of the largest lymph node, accuracy values of 66%, 69% and 73% were observed, respectively, whereas the ESGAR method was associated with an accuracy of 65%. The application of McNemar’s test showed no statistically significant differences. In the cohort of 39 patients, by applying the aforementioned cut-off values, the resulting accuracy was 66% for the short-axis, 77% for the long-axis, 74% for the volume and 67% for the number of lymph nodes (ESGAR method). Conclusion: The assessment of the volume or of the long-axis diameter of the largest visible lymph node after CRT in locally advanced rectal cancer is not inferior to the ESGAR method; therefore, it allows for a simpler and quicker evaluation of nodal status in these patients.File | Dimensione | Formato | |
---|---|---|---|
Tesi_Dejan Hajdukovic_PDF-A.pdf
accesso riservato
Dimensione
2.47 MB
Formato
Adobe PDF
|
2.47 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/40448