Prospective Assessment of Rectal Cancer Staging and Clinical Management with Both Radial and Linear EUS

Abstract

Background: Preoperative staging of rectal cancer is critical to providing appropriate care. EUS has a high diagnostic accuracy for rectal cancer T and N staging. In our experience, linear EUS is more sensitive than radial EUS for detecting malignant LNs while performance characteristics of both modalities appear comparable for T staging. A comparison of these two modalities has never been previously reported. Aims: Prospectively study pts with rectal cancer to assess: 1.) T/N stage accuracy of linear EUS when added to radial EUS, and 2.) whether altered T/N stage impacts pt management. Methods: 77 consecutive rectal cancer pts referred for EUS staging underwent radial followed by linear EUS. T and N stage were separately recorded for each modality. Suspicious LNs were sampled by FNA, except when sampling through the primary tumor was required. The gold standard for N stage was surgical pathology for N0 pts and surgical pathology and/or FNA results for N1 pts. Gold standard for T stage was surgical pathology. Results: The addition of linear EUS modified TNM staging for 17/77 (22%) pts. These findings altered clinical care in 10/77 (13%). In 3/77 pts, T stage differed and surgical pathology confirmed the linear exam findings in 2 pts. In 1 pt a benign appearing perirectal LN was seen only by radial EUS and confirmed as benign at surgery. In 1 pt a presumed malignant LN on radial exam was found by linear exam to be an asymmetric seminal vesicle and confirmed at surgery. In 11/76 pts LNs were detected only with linear EUS. Among them, 3/11 were benign and the TNM stage and management did not change. In 7/12 pts linear EUS alone identified malignant LNs (and confirmed by FNA) thereby altering management and indicating a need for preoperative chemoradiation. For 3/13 pts linear exam alone detected malignant iliac LNs (M1), which led to an expanded radiation field and precluded surgery. In the remaining pts the combined linear/radial exam did not result in a change in T, N, or M stage. Conclusions: These data demonstrate enhanced detection of malignant LNs in rectal cancer when linear EUS is combined with radial EUS exam. T stage accuracy appears equivalent for both imaging modalities. Overall, the addition of linear EUS changed tumor stage and/or clinical management in 15% of patients, supporting its adjunctive role in rectal cancer evaluation. Background: Preoperative staging of rectal cancer is critical to providing appropriate care. EUS has a high diagnostic accuracy for rectal cancer T and N staging. In our experience, linear EUS is more sensitive than radial EUS for detecting malignant LNs while performance characteristics of both modalities appear comparable for T staging. A comparison of these two modalities has never been previously reported. Aims: Prospectively study pts with rectal cancer to assess: 1.) T/N stage accuracy of linear EUS when added to radial EUS, and 2.) whether altered T/N stage impacts pt management. Methods: 77 consecutive rectal cancer pts referred for EUS staging underwent radial followed by linear EUS. T and N stage were separately recorded for each modality. Suspicious LNs were sampled by FNA, except when sampling through the primary tumor was required. The gold standard for N stage was surgical pathology for N0 pts and surgical pathology and/or FNA results for N1 pts. Gold standard for T stage was surgical pathology. Results: The addition of linear EUS modified TNM staging for 17/77 (22%) pts. These findings altered clinical care in 10/77 (13%). In 3/77 pts, T stage differed and surgical pathology confirmed the linear exam findings in 2 pts. In 1 pt a benign appearing perirectal LN was seen only by radial EUS and confirmed as benign at surgery. In 1 pt a presumed malignant LN on radial exam was found by linear exam to be an asymmetric seminal vesicle and confirmed at surgery. In 11/76 pts LNs were detected only with linear EUS. Among them, 3/11 were benign and the TNM stage and management did not change. In 7/12 pts linear EUS alone identified malignant LNs (and confirmed by FNA) thereby altering management and indicating a need for preoperative chemoradiation. For 3/13 pts linear exam alone detected malignant iliac LNs (M1), which led to an expanded radiation field and precluded surgery. In the remaining pts the combined linear/radial exam did not result in a change in T, N, or M stage. Conclusions: These data demonstrate enhanced detection of malignant LNs in rectal cancer when linear EUS is combined with radial EUS exam. T stage accuracy appears equivalent for both imaging modalities. Overall, the addition of linear EUS changed tumor stage and/or clinical management in 15% of patients, supporting its adjunctive role in rectal cancer evaluation.