Colorectal Surgery Wiki

Ciseł 2019 - POLISH II Long-Term Results

Properties1
tagscolorectal, rectal-cancer, tnt, neoadjuvant, radiotherapy, chemotherapy

4 min read

Ciseł 2019 - POLISH II Long-Term Results

Full citation: Ciseł B, Pietrzak L, Michalski W, Wyrwicz L, Rutkowski A, Kosakowska E, Cencelewicz A, Spałek M, Polkowski W, Jankiewicz M, et al., for the Polish Colorectal Study Group. Long-course preoperative chemoradiation versus 5 × 5 Gy and consolidation chemotherapy for clinical T4 and fixed clinical T3 rectal cancer: long-term results of the randomized Polish II study. Ann Oncol. 2019;30(8):1298-1303. ClinicalTrials.gov NCT00833131.

Raw PDF: raw/Ciseł 2019 - POLISH II Long-Term Results.pdf

Study type / design

Multicenter (39 Polish institutions), randomized, phase III, 2-arm trial. Patients with cT4 or palpably fixed cT3 rectal cancer (primarily or locally recurrent, involving adjacent organs, or a palpably fixed cT3 lesion — a higher-risk population than a typical resectable-LARC trial) were randomized to:

  • Short-course/CCT arm: 5 × 5 Gy radiotherapy over 5 days, followed by 3 cycles of FOLFOX4 (consolidation chemotherapy, CCT) before surgery — a short-course-radiotherapy-based TNT design, the same broad category as Bahadoer 2021 - RAPIDO Trial and Jin 2022 - STELLAR Trial.
  • Chemoradiation arm: long-course CRT, 50.4 Gy with concurrent bolus 5-FU, leucovorin, and oxaliplatin.

515 patients analyzed (261 short-course/CCT, 254 chemoradiation). This is the long-term update (median follow-up 7.0 years) of an earlier analysis (Bujko et al., Ann Oncol 2016) that had reported an early OS benefit for short-course/CCT at a median 3 years’ follow-up — the original 3-year primary publication is not independently ingested in this wiki; only this long-term update is.

Key findings

  • OS: no significant difference at long-term follow-up (HR 0.90, 95% CI 0.70-1.15, P=0.38). At 8 years, OS was 49% in both groups. The early OS advantage for short-course/CCT seen at 3 years (9% absolute difference, 95% CI 0.5%-17%) disappeared with longer follow-up — it did not represent a durable survival benefit.
  • DFS: no significant difference (HR 0.95, 95% CI 0.75-1.19, P=0.65). 8-year DFS 43% (short-course/CCT) vs 41% (chemoradiation).
  • Local failure (cumulative incidence, competing-risk analysis): no significant difference — HR 1.08, 95% CI 0.70-1.23, P=0.60; 35% vs 32%.
  • Distant metastases: no significant difference — HR 1.10, 95% CI 0.68-1.23, P=0.54; 36% vs 34%.
  • Late complications: similar rates (P=0.66); grade 3+ 11% (short-course/CCT) vs 9% (chemoradiation).
  • Radical resection rate (primary endpoint of the original trial, reported previously, restated here): 77% vs 71%, P=0.07 (not significant). pCR (ypT0N0) 16% vs 12%, P=0.17 (not significant).
  • Conclusion (authors’ own words): “The superiority of preoperative short-course/CCT over chemoradiation was not demonstrated.” This is a genuinely neutral/negative trial for the short-course-RT-based TNT strategy tested here — no efficacy benefit in either direction, and no locoregional-control trade-off either.

Limitations

  • Higher-risk population (cT4/fixed cT3) than most other TNT trials — not directly comparable to RAPIDO/PRODIGE 23/STELLAR’s broader LARC populations without caveat.
  • Pelvic MRI was not mandatory (only 66% of patients had it) — a real staging-quality limitation by modern standards.
  • 5-FU delivered as bolus in the chemoradiation arm rather than continuous infusion/capecitabine (the authors argue this is unlikely to change conclusions, citing two supporting trials, but it’s a design weakness relative to contemporary practice).
  • Oxaliplatin use in the chemoradiation arm changed mid-trial (per-institution discretion from 2012 onward after a separate trial showed no oxaliplatin benefit) — a protocol inconsistency.
  • No patient-reported outcomes / quality-of-life or functional (bowel, sexual) comparison — a real gap given this wiki’s broader interest in TNT’s late-toxicity trade-offs.
  • Discussion explicitly proposes short-course RT + delayed surgery without CCT as a worthwhile third arm that was missing from this trial — the true added value of consolidation chemotherapy itself (as opposed to short-course RT with delayed surgery generally) remains unproven per this paper’s own literature review.

Relevance

Adds a fourth primary-source data point to this wiki’s cross-trial TNT efficacy comparison (see Total Neoadjuvant Therapy (TNT)), alongside RAPIDO, PRODIGE 23, and STELLAR — previously known only secondhand via the Boublikova review’s table. Unlike RAPIDO (worse local control) or PRODIGE 23/STELLAR (better local control), POLISH II shows no difference in either direction for local control, distant metastases, or long-term OS — a genuinely neutral result, and a useful counterpoint against over-generalizing either “TNT trades local for distant control” (RAPIDO) or “TNT improves everything” (PRODIGE 23) from just the other three trials. Directly closes the “POLISH II is still not independently ingested” open item on Total Neoadjuvant Therapy (TNT) and Rectal Cancer.

Graph View

Backlinks