|
Back to 2017 Program and Abstracts
LOCAL MULTIVISCERAL RESECTION AFTER NEOADJUVANT TREATMENT FOR CLINICALLY LOCALLY ADVANCED RECTAL CANCER
Anne M. Dinaux*1,2, Lieve G. Leijssen1,2, Liliana G. Bordeianou1,2, Hiroko Kunitake1,2, David L. Berger1,2
1General and Gastrointestinal Surgery, Massachusetts General Hospital, Boston, MA; 2Harvard Medical School, Boston, MA
Introduction
Multivisceral resection is occasionally needed to obtain clear margins in patients with locally advanced rectal cancer. The literature on this subject demonstrates good oncological outcomes when clear margins are achieved. However, it is not uniform; most papers mix patients with recurrent and primary disease as well as patients who did not receive neoadjuvant with those who did. This study focuses solely on those patients who received neoadjuvant treatment for known locally advanced rectal cancers and underwent a local multivisceral R0-resection.
Methods
A retrospective analysis was performed on all patients with a transmural (cT3/cT4, based on preoperative imaging), primary rectal cancer with no baseline metastasis. All patients received an R0-resection at a single institute and were retrieved from a prospectively maintained database. The cohort was divided based on whether patients underwent a local multivisceral resection or not. Patients who had a distant multivisceral excision in the same procedure were excluded.
Results
A total of 279 patients were included, of whom 29 (10.4%) underwent a local multivisceral resection (LMVR). LMVR patients were significantly more often female (62.1% vs. non-LMVR 32.4%; P=0.002). Pre-neoadjuvant treatment staging scans showed a T4 tumor in 51.7% of the LMVR patients, compared to 5.7% of the non-LMVR patients (P<0.001) and clinical AJCC staging was similar (P=0.917). Operative duration and admission duration were significantly longer in the patients with LMVR, although 30-day complication rates were alike.
Besides a statistically significant higher high grade disease rate in the LMVR group, EMVI, Large vessel -, small vessel -, and perineural invasion showed trends towards poorer outcomes in the LMVR-group (see table).
Long-term oncological outcomes showed clinically significant differences in distant metastatic disease (LMVR 20.7% vs. non-LMVR 11.7%; P=0.23) and rectal cancer death rates (LMVR 13.8% vs. non-LMVR 6.1%; P=0.13). There was 1 local recurrence in the LMVR group (3.4%), compared to 13 in the non-LMVR group (5.3%). Comparing long-term outcomes pathologic stage for stage did not demonstrate statistically significant differences; however, small numbers might cause Type II errors. Median recurrent disease free survival and overall survival also showed no significant differences.
Conclusions
Despite the fact that all patients in this cohort received an R0-resection, the patients who underwent a local multivisceral resection after neoadjuvant treatment for a transmural tumor are at higher risk for distant disease recurrence and disease related death, compared to those who received neoadjuvant treatment for a transmural tumor but did not undergo a multivisceral resection.
Pathologic characteristics
| | No multivisceral resection | Local multivisceral resection | P-value | | n = | 250 | 29 | | | EMVI | 26 (10.5%) | 7 (24.1%) | 0.061 a | | High grade disease | 5 (5.7%) | 6 (20.7%) | 0.011 a | | Large vessel invasion | 30 (12.1%) | 6 (20.7%) | 0.413 | | Small vessel invasion | 29 (11.7%) | 7 (24.1%) | 0.158 | | Perineural invasion | 40 (16.2%) | 7 (24.1%) | 0.297 |
a: Fishers exact test used to determine statistical significance
Back to 2017 Program and Abstracts
|
