CROSS
Neoadjuvant chemoradiotherapy plus surgery versus surgery alone for resectable oesophageal or oesophagogastric-junction cancer (ChemoRadiotherapy for Oesophageal cancer followed by Surgery Study)
N = 366
- Population
- Clinically resectable oesophageal or oesophagogastric-junction cancer, clinical stage T1N1M0 or T2-3N0-1M0 (TNM sixth edition). Of 366 analysed: 75% adenocarcinoma, 23% squamous-cell carcinoma, 2% large-cell undifferentiated carcinoma
- Design
- Randomised controlled trial in eight Dutch centres (1:1, permuted blocks). Experimental arm: weekly carboplatin (AUC 2) and paclitaxel (50 mg/m2) for 5 weeks with concurrent radiotherapy (41.4 Gy in 23 fractions), then surgery; control arm: surgery alone. 368 patients enrolled from March 2004 to December 2008; 366 analysed. Registered with the Netherlands Trial Register, NTR487 (not an NCT number).
- Primary endpoint
- OS
- Arms
- Chemoradiotherapy then surgery (experimental, n = 178): Weekly carboplatin (AUC 2) and paclitaxel (50 mg/m2) for 5 weeks with concurrent radiotherapy (41.4 Gy in 23 fractions), followed by surgery
- Surgery alone (control, n = 188): Surgery alone
- People
- van Hagen P, CROSS Study Group
OSHR 0.657 (95% CI 0.495–0.871)Δ median OS+25.4 monthsΔ R0 resection+23 pointsOSHR 0.68 (95% CI 0.53–0.88)Δ median OS+24.6 monthsOSHR 0.48 (95% CI 0.28–0.83)
Show the numbers
| Endpoint | Analysis | Chemoradiotherapy then surgery | Surgery alone | Effect |
|---|---|---|---|---|
| OS366 analysed patients (178 vs 188) | primaryvan 2012 | median 49.4 months | median 24 months | HR 0.657 (95% CI 0.495–0.871)P=0.003Initial report. The abstract gives no median follow-up or confidence intervals for the medians; the long-term Lancet Oncology abstract states a median follow-up of 45 months for this first report. |
| R0 resectionPatients in each group, as reported in the abstract (denominators not stated there) | primaryvan 2012 | 92% | 69% | P<0.001R0 means complete resection with no tumour within 1 mm of the resection margins. |
| pCRPatients who underwent resection after chemoradiotherapy (47 of 161) | primaryvan 2012 | 29% | — | The abstract reports no pathological response figure for the surgery-alone group. Postoperative complications were similar between groups and in-hospital mortality was 4% in both. |
| OSIntention-to-treat population, 366 analysed patients (178 vs 188) | long term-updatemedian follow-up 84.1 monthsShapiro 2015 | median 48.6 months (32.1–65.1) | median 24 months (14.2–33.7) | HR 0.68 (95% CI 0.53–0.88)P=0.003Minimum follow-up 5 years; no adverse-event data were collected beyond the initial report. |
| OSSquamous-cell carcinoma subgroup | long term-updateShapiro 2015 | median 81.6 months (47.2–116.0) | median 21.1 months (15.4–26.7) | HR 0.48 (95% CI 0.28–0.83)P=0.008Subgroup analysis by histology. |
| OSAdenocarcinoma subgroup | long term-updateShapiro 2015 | median 43.2 months (24.9–61.4) | median 27.1 months (13.0–41.2) | HR 0.73 (95% CI 0.55–0.98)P=0.038Subgroup analysis by histology. |
| OS366 randomly assigned patients | long term-updatemedian follow-up 147 monthsEyck 2021 | 38% at 10 y | 25% at 10 y | HR 0.70 (95% CI 0.55–0.89)The abstract states the effect on overall survival was not time-dependent (P for interaction 0.73). Death from oesophageal cancer: HR 0.60 (95% CI 0.46 to 0.80); death from other causes: HR 1.17 (95% CI 0.68 to 1.99). The 38% and 25% values are read as the chemoradiotherapy and surgery-alone arms respectively. |
| Endpoint | Analysis | Chemoradiotherapy then surgery | Surgery alone | Effect |
|---|---|---|---|---|
| OS366 analysed patients (178 vs 188) | primaryvan 2012 | median 49.4 months | median 24 months | HR 0.657 (95% CI 0.495–0.871)P=0.003Initial report. The abstract gives no median follow-up or confidence intervals for the medians; the long-term Lancet Oncology abstract states a median follow-up of 45 months for this first report. |
| R0 resectionPatients in each group, as reported in the abstract (denominators not stated there) | primaryvan 2012 | 92% | 69% | P<0.001R0 means complete resection with no tumour within 1 mm of the resection margins. |
| pCRPatients who underwent resection after chemoradiotherapy (47 of 161) | primaryvan 2012 | 29% | — | The abstract reports no pathological response figure for the surgery-alone group. Postoperative complications were similar between groups and in-hospital mortality was 4% in both. |
| OSIntention-to-treat population, 366 analysed patients (178 vs 188) | long term-updatemedian follow-up 84.1 monthsShapiro 2015 | median 48.6 months (32.1–65.1) | median 24 months (14.2–33.7) | HR 0.68 (95% CI 0.53–0.88)P=0.003Minimum follow-up 5 years; no adverse-event data were collected beyond the initial report. |
| OSSquamous-cell carcinoma subgroup | long term-updateShapiro 2015 | median 81.6 months (47.2–116.0) | median 21.1 months (15.4–26.7) | HR 0.48 (95% CI 0.28–0.83)P=0.008Subgroup analysis by histology. |
| OSAdenocarcinoma subgroup | long term-updateShapiro 2015 | median 43.2 months (24.9–61.4) | median 27.1 months (13.0–41.2) | HR 0.73 (95% CI 0.55–0.98)P=0.038Subgroup analysis by histology. |
| OS366 randomly assigned patients | long term-updatemedian follow-up 147 monthsEyck 2021 | 38% at 10 y | 25% at 10 y | HR 0.70 (95% CI 0.55–0.89)The abstract states the effect on overall survival was not time-dependent (P for interaction 0.73). Death from oesophageal cancer: HR 0.60 (95% CI 0.46 to 0.80); death from other causes: HR 1.17 (95% CI 0.68 to 1.99). The 38% and 25% values are read as the chemoradiotherapy and surgery-alone arms respectively. |
Data table
| Endpoint | HR | 95% CI |
|---|---|---|
| OS | 0.657 | 0.495–0.871 |
| OS | 0.68 | 0.53–0.88 |
| OS | 0.48 | 0.28–0.83 |
| OS | 0.73 | 0.55–0.98 |
| OS | 0.7 | 0.55–0.89 |
- Chemoradiotherapy then surgery (experimental)
- Surgery alone (control)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Chemoradiotherapy then surgery | Surgery alone |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 31 | 64.7% | 40.8% |
| 62 | 41.9% | 16.7% |
| 93 | 27.1% | 6.8% |
| 124 | 17.6% | 2.8% |
- Chemoradiotherapy then surgery (experimental)
- Surgery alone (control)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Chemoradiotherapy then surgery | Surgery alone |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 30 | 65.2% | 42.0% |
| 61 | 41.9% | 17.2% |
| 91 | 27.3% | 7.2% |
| 122 | 17.6% | 2.9% |
Appears in
Sources
van Hagen P, Hulshof MC, van Lanschot JJ, Steyerberg EW, van Berge Henegouwen MI, Wijnhoven BP, et al. Preoperative chemoradiotherapy for esophageal or junctional cancer. N Engl J Med. 2012;366(22):2074-84.
Shapiro J, van Lanschot JJB, Hulshof MCCM, van Hagen P, van Berge Henegouwen MI, Wijnhoven BPL, et al. Neoadjuvant chemoradiotherapy plus surgery versus surgery alone for oesophageal or junctional cancer (CROSS): long-term results of a randomised controlled trial. Lancet Oncol. 2015;16(9):1090-1098.
Eyck BM, van Lanschot JJB, Hulshof MCCM, van der Wilk BJ, Shapiro J, van Hagen P, et al. Ten-Year Outcome of Neoadjuvant Chemoradiotherapy Plus Surgery for Esophageal Cancer: The Randomized Controlled CROSS Trial. J Clin Oncol. 2021;39(18):1995-2004.