Boublikova 2023 — TNT in Rectal Cancer Review
Citation: Boublikova L, Novakova A, Simsa J, Lohynska R. Total neoadjuvant therapy in rectal cancer: the evidence and expectations. Crit Rev Oncol Hematol. 2023;192:104196. doi:10.1016/j.critrevonc.2023.104196.
Type of source
Narrative review (not a primary trial) synthesizing four phase III RCTs — POLISH II, RAPIDO, PRODIGE 23, STELLAR — plus retrospective/phase II data (CAO/ARO/AIO-12), and giving practical guidance on indications, regimen choice, sequencing, and non-operative management (NOM).
The four phase III trials (per the review’s Table 1)
| Trial | Design | n | Experimental arm | Control arm | pCR (exp vs ctrl) | 3y-DFS | 3y-OS |
|---|---|---|---|---|---|---|---|
| POLISH II (Bujko 2016) | ↑R0 resection | 261/254 | SCRT → 3xFOLFOX4 → surgery | CRT+OXPw → surgery | 16% vs 12% | 53% vs 52% | 73% vs 65% |
| RAPIDO (Bahadoer 2020) | ↓distant treatment failure | 462/450 | SCRT → 6xCAPOX/9xFOLFOX4 → surgery | CRT → surgery ± adjuvant CAPOX/FOLFOX4 | 28% vs 14% | ~76% vs ~70% | 89% vs 89% |
| PRODIGE 23 (Conroy 2021) | ↑DFS | 231/230 | 6xmFOLFOXIRI → CRT → surgery → 6xmFOLFOX6/4xCAPE | CRT → surgery → 12xmFOLFOX/8xCAPE | 28% vs 13% | 76% vs 69% | 91% vs 88% (7y: 82% vs 76%) |
| STELLAR (Jin 2022) | non-inferiority DFS | 298/293 | SCRT → 4xCAPOX → surgery → 2xCAPOX | CRT → surgery → 6xCAPOX | 17% vs 12% | 65% vs 62% | 87% vs 75% |
(SCRT = short-course RT, 5×5Gy; CRT = long-course chemoradiotherapy, 50Gy + capecitabine/5FU)
Key findings
- TNT reduces distant metastases and increases DFS/MFS by 5-10% at 3 years; PRODIGE 23’s 7-year update showed a real OS benefit (82% vs 76%) — none of the trials individually was powered for OS given short follow-up, but longer follow-up is now bearing it out.
- pCR roughly doubled (to >25% in most trials) — the key enabler of non-operative management (NOM), though pCR and successful NOM are not perfectly correlated (OPRA trial’s 3y TME-free survival exceeded PRODIGE 23’s pCR rate by >10 points).
- Important trade-off in RAPIDO: locoregional recurrence was numerically higher with SCRT-based TNT from the start (6.3% vs 3.8%) and became significantly higher at longer follow-up (10% vs 6%). Possible explanations explored (radiotherapy technique, TME quality) were inconclusive. This tempers the “TNT is a clean win” framing — it’s a trade of distant control for a small increase in local recurrence risk, at least with the SCRT-based regimen.
- Toxicity: no significant increase in serious adverse events overall (20-40% across trials, similar treatment-related death rates 0.5-5%). Compliance much better in the neoadjuvant setting (80-90% completed) vs. adjuvant (25-50%) — a major practical argument for TNT regardless of survival numbers.
- Indications are risk-stratified, not blanket: clear benefit for T4, N2, MRF+, or EMVI+ tumors; T3N0 generally doesn’t benefit from TNT specifically; distal T3N0 tumors are the best NOM candidates (organ preservation has the highest payoff there).
- MMRd/MSI tumors (~5% of rectal cancers) are a distinct pathway: immunotherapy alone (PD-1/PD-L1 inhibitors — ipilimumab+nivolumab, dostarlimab, etc.) achieves 60-100% pCR, even without radiotherapy. Chemotherapy-based TNT is not recommended for this subgroup.
- NOM practicalities: candidates assessed 4-8 weeks post-TNT via DRE, endoscopy, MRI (MSK three-tiered response scheme). Repeat biopsy not recommended (false pos/neg risk, healing issues). Locoregional relapse on surveillance ~20-40%, mostly salvageable by surgery without DFS penalty; distant metastatic risk while on surveillance 5-15%, mostly associated with locoregional recurrence — a real risk that must be discussed with patients.
- Adjuvant chemo after TNT + surgery is not indicated if the full neoadjuvant course (3-4 months) was completed; exceptions: R1-2 resection, or incomplete neoadjuvant course.
Relevance
First rectal cancer / oncology source in the wiki (previous sources were diverticulitis surgery). Establishes Total Neoadjuvant Therapy (TNT) as a concept page and seeds a Rectal Cancer condition stub. Good candidate for cross-checking against future primary publications of RAPIDO, PRODIGE 23, STELLAR, or newer trials (e.g. JANUS, ACO/ARO/AIO-18) if ingested later — this review’s Table 1 numbers should be treated as a secondary summary, not a substitute for the primary papers if fine-grained detail is ever needed.