Local Excision vs Radical Resection of Colorectal Carcinoma
Local Excision vs Radical Resection of Colorectal Carcinoma
It has been interesting, when we looked at the national cancer database, the use of local excision, despite reviews such as yours that have shown worse outcomes, has tripled in the last 10 years in the United States. My question to you is whether the adverse outcomes in your study were predominantly related to rectal cancer or colon cancer.And in the cases of either,whatwas the use of adjuvant therapy in that group?
To take your second question first, unfortunately, SEER does not code for the use of adjuvant chemotherapy, sowe could not include that data in our analysis. A secondary Medicare analysis may be able to do that.
Regarding your first point on the studies of rectal cancer, I think one of the problems is, when Professor Tekkis and I set out late one night to do this analysis, with population-level data; the number of subgroups you analyze increases your chance of, really, type 1 errors. So, we agreed to limit this analysis to the subgroups of T1 and T2 tumors. We have adjusted all the analyses for rectal and colonic lesions, and further analysis would require breaking that down into 2 separate subgroups and repeating the models, which we could possibly do next. We have to be careful, as we pointed out, about identifying statistical abnormalities, which are not clinically.
Discussants: E.R. Sigurdson (Philadelphia, PA)
E.R. Sigurdson (Philadelphia, PA)
It has been interesting, when we looked at the national cancer database, the use of local excision, despite reviews such as yours that have shown worse outcomes, has tripled in the last 10 years in the United States. My question to you is whether the adverse outcomes in your study were predominantly related to rectal cancer or colon cancer.And in the cases of either,whatwas the use of adjuvant therapy in that group?
Response From A. Bhangu (London, UK)
To take your second question first, unfortunately, SEER does not code for the use of adjuvant chemotherapy, sowe could not include that data in our analysis. A secondary Medicare analysis may be able to do that.
Regarding your first point on the studies of rectal cancer, I think one of the problems is, when Professor Tekkis and I set out late one night to do this analysis, with population-level data; the number of subgroups you analyze increases your chance of, really, type 1 errors. So, we agreed to limit this analysis to the subgroups of T1 and T2 tumors. We have adjusted all the analyses for rectal and colonic lesions, and further analysis would require breaking that down into 2 separate subgroups and repeating the models, which we could possibly do next. We have to be careful, as we pointed out, about identifying statistical abnormalities, which are not clinically.
Source...