Neo-AEGIS
Trimodality therapy versus perioperative chemotherapy in the management of locally advanced adenocarcinoma of the oesophagus and oesophagogastric junction (Neo-AEGIS): an open-label, randomised, phase 3 trial
Phase 3N = 377NCT01726452
- Population
- Adenocarcinoma of the oesophagus or oesophagogastric junction, clinical tumour stage T2-3, nodal stage N0-3, M0 (Siewert types I to III for junctional tumours)
- Design
- Open-label randomised phase 3 trial at 24 centres in Europe (CTRIAL-IE 10-14). Perioperative chemotherapy (modified MAGIC-type epirubicin, platinum and fluoropyrimidine regimens before 2018; FLOT optional from 2018; 15% of that group received FLOT per the full text) versus trimodality therapy (CROSS regimen: 41.4 Gy in 23 fractions with carboplatin and paclitaxel, then surgery). 377 patients randomised (189 vs 188) between January 2013 and December 2020; 362 started treatment and form the intention-to-treat population. Originally powered for superiority of trimodality therapy; redesigned as non-inferiority (5% margin for perioperative chemotherapy) after FLOT became an option. Closed prematurely after a second interim futility analysis (143 deaths).
- Primary endpoint
- OS
- Arms
- Perioperative chemotherapy (control, n = 184): Three preoperative and three postoperative cycles of epirubicin plus cisplatin or oxaliplatin plus fluorouracil or capecitabine; FLOT (four preoperative and four postoperative cycles) as an option from 2018
- Trimodality therapy (CROSS regimen) (experimental, n = 178): 41.4 Gy in 23 fractions with carboplatin (AUC 2) and paclitaxel (50 mg/m2), then surgery
- People
- Reynolds JV
OSHR 1.03 (95% CI 0.77–1.38)Δ median OS+1.2 monthsΔ 36-month OS+2 pointsDFSHR 0.89 (95% CI 0.68–1.17)Δ median DFS−8.4 monthsΔ pCR+8 points
Show the numbers
| Endpoint | Analysis | Trimodality therapy (CROSS regimen) | Perioperative chemotherapy | Effect |
|---|---|---|---|---|
| OSIntention-to-treat population (362 patients who started treatment: 184 vs 178) | primarymedian follow-up 38.8 monthsReynolds 2023 | median 49.2 months (34.8–74.4); 57% at 3 y (49–64%) | median 48 months (33.6–64.8); 55% at 3 y (47–62%) | HR 1.03 (95% CI 0.77–1.38)P=0.82Hazard ratio direction: the full text defines the non-inferiority hazard ratio (1.16) with perioperative chemotherapy as the numerator and compares the upper confidence limit of 1.38 with it, so 1.03 is perioperative chemotherapy relative to trimodality therapy. At closure non-inferiority of perioperative chemotherapy was not established (upper bound 1.38 exceeded 1.16) and futility was not established either. The authors call the trial underpowered and incomplete; operative mortality 3% (5 deaths) vs 2% (4 deaths). |
| DFSIntention-to-treat population (184 vs 178) | primaryReynolds 2023 | median 24 months (18.0–40.8) | median 32.4 months (22.8–64.8) | HR 0.89 (95% CI 0.68–1.17)P=0.41Secondary endpoint; no significant difference. |
| pCRPatients with a pathology specimen (162 perioperative chemotherapy vs 167 trimodality) | primaryReynolds 2023 | 12% | 4% | P=0.012Secondary endpoint. Abstract: pathological complete response odds ratio 0.33 (95% CI 0.14 to 0.81), major pathological response 0.21 (0.12 to 0.38), R0 resection 0.21 (0.08 to 0.53), all favouring trimodality therapy. |
| Endpoint | Analysis | Trimodality therapy (CROSS regimen) | Perioperative chemotherapy | Effect |
|---|---|---|---|---|
| OSIntention-to-treat population (362 patients who started treatment: 184 vs 178) | primarymedian follow-up 38.8 monthsReynolds 2023 | median 49.2 months (34.8–74.4); 57% at 3 y (49–64%) | median 48 months (33.6–64.8); 55% at 3 y (47–62%) | HR 1.03 (95% CI 0.77–1.38)P=0.82Hazard ratio direction: the full text defines the non-inferiority hazard ratio (1.16) with perioperative chemotherapy as the numerator and compares the upper confidence limit of 1.38 with it, so 1.03 is perioperative chemotherapy relative to trimodality therapy. At closure non-inferiority of perioperative chemotherapy was not established (upper bound 1.38 exceeded 1.16) and futility was not established either. The authors call the trial underpowered and incomplete; operative mortality 3% (5 deaths) vs 2% (4 deaths). |
| DFSIntention-to-treat population (184 vs 178) | primaryReynolds 2023 | median 24 months (18.0–40.8) | median 32.4 months (22.8–64.8) | HR 0.89 (95% CI 0.68–1.17)P=0.41Secondary endpoint; no significant difference. |
| pCRPatients with a pathology specimen (162 perioperative chemotherapy vs 167 trimodality) | primaryReynolds 2023 | 12% | 4% | P=0.012Secondary endpoint. Abstract: pathological complete response odds ratio 0.33 (95% CI 0.14 to 0.81), major pathological response 0.21 (0.12 to 0.38), R0 resection 0.21 (0.08 to 0.53), all favouring trimodality therapy. |
Data table
| Endpoint | HR | 95% CI |
|---|---|---|
| OS | 1.03 | 0.77–1.38 |
| DFS | 0.89 | 0.68–1.17 |
- Perioperative chemotherapy (control)
- Trimodality therapy (CROSS regimen) (experimental)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Perioperative chemotherapy | Trimodality therapy (CROSS regimen) |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 31 | 63.9% | 64.6% |
| 62 | 40.8% | 41.7% |
| 92 | 26.5% | 27.4% |
| 123 | 16.9% | 17.7% |
- Perioperative chemotherapy (control)
- Trimodality therapy (CROSS regimen) (experimental)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Perioperative chemotherapy | Trimodality therapy (CROSS regimen) |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 20 | 65.2% | 56.1% |
| 41 | 41.6% | 30.6% |
| 61 | 27.1% | 17.2% |
| 81 | 17.7% | 9.6% |
Appears in
Sources
Reynolds JV, Preston SR, O'Neill B, Lowery MA, Baeksgaard L, Crosby T, et al. Trimodality therapy versus perioperative chemotherapy in the management of locally advanced adenocarcinoma of the oesophagus and oesophagogastric junction (Neo-AEGIS): an open-label, randomised, phase 3 trial. Lancet Gastroenterol Hepatol. 2023;8(11):1015-1027.