Rectal cancer

Prognostic factors
Location (distal is worse)
T/N/M stage
Grade
Signet cell carcinoma (bad)
LVSI
Circumferential/obstructing tumors
Fixed tumor

Imaging
EUS and MRI improve staging accuracy but both lose accuracy after neo-adjuvant treatment

Surgery
Total mesorectal excision (TME)
-Remove involved rectum and adjacent mesorectum along pelvic fascial planes
-2cm longitudinal margins considered adequate
-12 to 15 nodes considered adequate nodal dissection
-Radial margin status is a strong predictor for local control

Adjuvant trials
NSABP R-01: observation vs. chemoRT; chemoRT improved DFS and local control without impacting overall survival

NSABP R-02: chemo only vs. chemoRT: adding RT improved local control without changing survival

GITSG: observation vs. chemo only vs. RT only vs. chemoRT: ChemoRT improved OS, DFS, local control compared with observation arm

NCCTG: RT only vs. chemoRT: adding chemo improved OS, DFS, and local control

Preop trials
Swedish: 25Gy in 5fx followed by surgery vs. surgery alone (PMID 16110023)
RT improved local control (75 vs 88%) and overall survival (9-yr 48 vs 58%)
OS improvement may have been affected by the fact that TME was NOT performed in this study

Dutch: 25Gy in 5fx followed by TME vs. TME alone (PMID 10391155)
RT did not change OS but did improve local control; higher incidence of perineal complications in the RT arm

German: compared preop vs. postop chemoRT (PMID 15496622)
Preoperative chemoRT was associated with improved local control and improved toxicity without affecting overall survival

MRC/NCIC: (PMID 19269519) compared preop 5 Gy x 5 vs. postop 45 Gy chemo (patients with positive margins only)
Preoperative treatment was associated with significant improvements in local control and DFS, no effect on OS

TROG 0104 (PMID 2300801): compared preop short-course (5 Gy x 5) RT only vs. preop long course chemo-RT (45 Gy + capecitabine) in patients with T3N0-2 rectal cancer. 3-year overall survival rates were 74% and 70% for short-course and long course respectively (p-NS); local recurrence rates were 7.5% and 4.4% for short- and long-course respectively (p-NS) with much of the difference in local recurrence appearing to be driven by higher recurrence risk in low lying tumors receiving short course treatment; caveat of very small n's (12 local recurrences in short course group with 6 of these being in low lying lesions; 7 local recurrences in long course group with one of these in a low lying tumor). Similar late toxicity rates.