Rectal cancer

Overview

• Definition: rectum starts at anal verge and extends ~15 cm proximally (or to rectosigmoid junction).
• Distance from anal verge on rigid sigmoidoscopy:
  • Upper rectum (>10 cm): behaves like colon.
  • Mid-rectum (5 to 10 cm): trimodality candidate.
  • Lower rectum (<5 cm): risk of APR (loss of sphincters).
• Key difference from colon cancer: rectal cancers have a higher risk of local recurrence because of the anatomic difficulty in achieving clear circumferential resection margins (CRM). Historically trimodality (chemo, RT, surgery), but treatment is now risk- and biomarker-adapted: TNT for many locally advanced pMMR tumors, selected patients omit RT (PROSPECT), cCR allows nonoperative watch-and-wait, and dMMR/MSI-H tumors can achieve durable cCR with immunotherapy alone.
• Outcomes by modality (5-yr OS, historical): surgery alone ~45%; + adjuvant RT ~50% (+~5%); + adjuvant chemo ~50% (+~5%); + chemoRT ~60% (+~10 to 15%). Because chemoRT was so beneficial, it was moved to the neoadjuvant setting to downstage, improve sphincter preservation, deliver chemo before anatomic disruption, and reduce toxicity.

Workup / staging

• Pelvic MRI with rectal protocol: defines T-stage, mesorectal fascia (CRM), N status, lateral pelvic LN. Now the preferred staging approach (more precise T stage and CRM assessment than EUS).
• Endorectal ultrasound (EUS): complementary, especially for early T1/T2; EUS cannot replace MRI's CRM and nodal assessment, and is best reserved for early T1 vs T2 or when MRI is contraindicated.
• CT chest/abdomen/pelvis for distant staging. No role for routine PET in rectal cancer staging.
• Mandatory biomarkers: MMR/MSI on biopsy. RAS/BRAF/HER2 if metastatic.
• CEA baseline.

Surgical approaches

• Total mesorectal excision (TME): standard technique; removes rectum with mesorectum up to the levators, plus perirectal nodes and lymphatics harboring microscopic disease. Honors tissue planes and reduces local seeding. With pre- or postoperative CRT, 5-yr local recurrence <5 to 10%.
• Low anterior resection (LAR) / high anterior resection (HAR): sphincter-preserving.
• Abdominoperineal resection (APR): resects rectum and anus (plus sigmoid), permanent end colostomy; not sphincter-preserving; historically SOC for low rectal cancer.
• Local excision (TEM/TAMIS): restricted to T1 without high-risk factors (and T1 sm1 superficial submucosal).

Localized rectal cancer

• LAR or APR with TME. cT2N0M0: proceed directly to TME, no benefit from neoadjuvant chemo or chemoRT.
• T1 sm1 (superficial submucosal): transanal local excision (TEM/TAMIS) is an option.

• German rectal trial (Sauer et al): neoadjuvant chemoRT → surgery vs surgery → adjuvant chemoRT. Preoperative chemoRT improved local recurrence (~6% vs 13% at 5 yr; also cited 7% vs 11%), ↑ sphincter preservation only in the subgroup pre-judged to need APR (39% vs 19%); overall rates not significantly different, ↓ chronic anastomotic stenosis (4% vs 12%), pCR ~8%, with less toxicity but no OS advantage. Established neoadjuvant CRT with 5-FU as radiosensitizer as SOC for stage II/III.
• RT options: long-course chemoRT to 50.4 Gy (45 + 5.4 Gy boost) over 5 to 6 weeks with concurrent infusional 5-FU or capecitabine (825 mg/m² BID on RT days; two phase III trials showed noninferiority to infusional 5-FU), then surgery ~8 to 12 weeks after RT; OR short-course RT (5 × 5 Gy). Long-course CRT downstages; short-course RT also downstages when surgery is delayed (Stockholm III) or followed by consolidation chemo (RAPIDO, pCR 28%). Only short-course RT with immediate surgery leaves little time for regression; both reduce locoregional failure. Two phase III trials found short-course and long-course similar for local recurrence and OS.
• Oxaliplatin as radiosensitizer: Italian, French, and US phase III trials showed adding oxaliplatin to fluoropyrimidine during CRT did NOT significantly increase pCR but significantly increased diarrhea; not recommended outside a trial (both CAO/ARO/AIO-04 and FOWARC showed higher pCR with oxaliplatin plus RT (FOWARC 27.5% vs 14.0%), but neither justifies routine oxaliplatin as a radiosensitizer).
• Biologics during CRT: bevacizumab, cetuximab, and panitumumab should NOT be added to the CRT backbone (wound/anastomotic complications with bev; disappointing pCR with anti-EGFR).

• Concept: deliver both chemotherapy and RT before TME (in either order), improving compliance (postoperative chemo completion is often ≤50%), increasing pCR, and enabling organ preservation.
• RAPIDO (Bahadoer Lancet Oncol 2021): short-course RT (5 × 5) → consolidation CAPOX/FOLFOX → surgery vs chemoRT → surgery → optional adjuvant chemo (CAPOX/FOLFOX per institutional policy). Lower disease-related treatment failure in the TNT arm (~23.7% vs 30.4%), ↑ pCR (28% vs 14%), same OS (5-yr ~80%). Note: 5-yr update (2023) showed ↑ local recurrence → careful patient selection now recommended.
• PRODIGE 23 (induction mFOLFIRINOX): improved DFS, metastasis-free survival, and OS on long-term follow-up (7-yr OS 81.9% vs 76.1%, p=0.033).
• OPRA: compared induction chemo then chemoRT vs chemoRT then consolidation chemo; showed consolidation sequencing (chemoRT then FOLFOX or CAPOX) improved organ preservation (5-yr TME-free survival ~54% vs 39%) with no DFS difference; supports watch-and-wait by clinical CR, not pCR. ~50% organ preservation at 3 yr with a cCR-driven watch-and-wait approach.
• Standard timing: after long-course CRT without consolidation, surgery ~8 to 12 wk after RT. With TNT, surgery follows completion of the entire regimen and response assessment (often well beyond 12 wk after RT); OPRA restaged ~8 wk after completing TNT.
• NRG GI002: failed to show benefit of adding neoadjuvant pembrolizumab (in an unselected population).

• PROSPECT (N1048) (Schrag NEJM 2023; PMID 37272512): for stage II to III with lower-risk features (T2N+ or T3N0/N+, no threatened CRM), neoadjuvant FOLFOX × 6 cycles → response-driven omission of pelvic chemoRT (only ~10% needed RT). Non-inferior 5-yr DFS vs chemoRT + 5-FU; spares RT-related toxicity (sexual dysfunction, fertility, bowel). Practice-changing for selected patients. (Earlier FOWARC in China similarly compared FOLFOX alone with FOLFOX/RT or 5-FU/RT and found similar local recurrence, 3-yr DFS, and OS.)

• Eligible: clinical complete response (cCR) after neoadjuvant treatment: negative DRE, MRI, and sigmoidoscopy.
• Evidence: Brazilian (Habr-Gama) data and a meta-analysis of 23 studies (867 pts) plus an international registry show high organ preservation with local regrowth ~15 to 25% (higher, ~30 to 40%, in prospective TNT cohorts such as OPRA), most salvageable.
• Surveillance: intensive (DRE + sigmoidoscopy ~q3 mo, MRI q6 mo). Regrowth ~15 to 25% overall, ~30 to 40% after TNT, mostly within 2 yr → salvage TME at experienced centers.
• Caveat: not the universal SOC; should be done in the context of a trial or at experienced multidisciplinary centers per NCCN, after careful patient discussion.

• Cercek/MSKCC dostarlimab study (Cercek NEJM 2022, PMID 35660797, initial 12 pts, 100% cCR; ASCO 2024 update (JCO abstract), 42 pts, 100% cCR, median follow-up 26 mo in the 24 pts with sustained cCR for at least 12 mo; NEJM 2025 pan-tumor (PMID 40293177): all 49 rectal pts cCR; nonrectal dMMR early-stage tumors 65% cCR (~82% overall)).
• Regimen: dostarlimab 500 mg IV q3wk × 9 doses (6 mo).
• Outcome: 100% cCR in the dMMR rectal cohort allows surgery, chemo, and RT to be AVOIDED in responders; no progression reported during treatment.
• Status: not yet FDA-approved specifically for this indication (dostarlimab is FDA-approved for dMMR endometrial 2021, expanded 2023). NCCN lists it as an option for dMMR locally advanced rectal ca; offer at experienced centers with structured follow-up.
• Caveat: all 49 rectal pts had cCR, but cohorts are small with limited follow-up (2-yr RFS 92% across all 117 pts) and nonrectal dMMR tumors respond less often (65% cCR); MMR/MSI confirmation and intensive surveillance are essential. Salvage TME remains feasible.

Adjuvant therapy

• After TNT + surgery: complete remaining chemo if not delivered preoperatively.
• After upfront surgery (found to be stage II/III): postoperative chemoRT + chemo (older paradigm). ADORE (Korea): after neoadjuvant CRT, adjuvant FOLFOX beat bolus 5-FU (6-yr DFS 68.2% vs 56.8%, HR 0.63). NCCN recommends adjuvant chemo even after a pathologic CR (not universally accepted).
• All stage II/III should receive systemic therapy (FOLFOX or CAPOX) either before (TNT) or after surgery. No oxaliplatin benefit in the neoadjuvant CRT phase; capecitabine may substitute for 5-FU.

Metastatic rectal cancer

• Same systemic tx as colon ca (see CRC note: FOLFOX/FOLFIRI ± bev or anti-EGFR by RAS/BRAF and sidedness; targeted options for BRAF V600E, HER2, KRAS G12C; IO for dMMR).
• Local therapy: palliative RT, diverting colostomy, stent for obstruction.
• Liver/lung metastasectomy for oligometastatic disease.

High-yield rectal pearls

• Pelvic MRI is the preferred staging tool (CRM, T/N); no routine PET.
• TME is the standard surgery; APR for low tumors (sphincter loss), LAR/HAR preserve sphincter.
• cT2N0M0: TME directly, no neoadjuvant therapy.
• TNT preferred over standard neoadjuvant chemoRT, then TME, then adjuvant chemo for locally advanced stage II to III (RAPIDO, PRODIGE 23).
• PROSPECT: neoadjuvant FOLFOX × 6 can omit pelvic RT in selected lower-risk stage II/III.
• Watch-and-wait for cCR after TNT (OPRA); ~30 to 40% regrow, mostly salvageable.
• dMMR rectal: dostarlimab × 6 mo → 100% cCR (Cercek). Game-changer.
• Do NOT add oxaliplatin, bevacizumab, cetuximab, or panitumumab to the CRT backbone.
• CRM threatened on MRI → benefits from RT + chemo. APR avoidance is a goal; TNT often enables sphincter preservation.