New England Journal of Medicine May 31, 2012

Preoperative Chemoradiotherapy for Esophageal or Junctional Cancer (CROSS Trial)

P. van Hagen, M.C.C.M. Hulshof, J.J.B. van Lanschot, E.W. Steyerberg, M.I. van Berge Henegouwen, B.P.L. Wijnhoven et al.

Bottom Line

Neoadjuvant chemoradiotherapy with weekly carboplatin and paclitaxel followed by surgery significantly improved overall survival compared to surgery alone in patients with potentially curable esophageal or esophagogastric-junction cancer.

Key Findings

1. Median overall survival was significantly prolonged in the chemoradiotherapy-surgery group compared to the surgery alone group (49.4 months vs. 24.0 months; HR 0.657, 95% CI 0.495-0.871, p=0.003) [2.1.3].
2. Complete resection with negative margins (R0) was achieved in 92% of the chemoradiotherapy group versus 69% in the surgery alone group (p<0.001).
3. A pathological complete response (pCR) was observed in 29% of patients who received chemoradiotherapy (49% in squamous cell carcinoma, 23% in adenocarcinoma).
4. Postoperative complications and in-hospital mortality (4% in both arms) were similar between the two groups, and 95% of patients completed the entire chemoradiotherapy protocol.

Study Design

Design
Randomized Controlled Trial
Open-Label
Sample
366
Patients
Duration
45.4 mo
Median
Setting
Multicenter, Netherlands
Population Patients with potentially curable, locally advanced esophageal or esophagogastric-junction cancer (cT1N1M0 or cT2-3N0-1M0); 75% adenocarcinoma, 23% squamous-cell carcinoma.
Intervention Preoperative chemoradiotherapy: 5 weekly cycles of carboplatin (AUC 2) and paclitaxel (50 mg/m2) with concurrent radiotherapy (41.4 Gy in 23 fractions), followed by surgery.
Comparator Surgery alone.
Outcome Overall survival.

Study Limitations

The trial was open-label, which is inherent to surgical versus non-surgical initial treatment designs but can introduce bias.
The radiation dose of 41.4 Gy is lower than the definitive dose of 50.4 Gy, leaving uncertainty about the optimal dose for patients who do not proceed to surgery.
The study compared chemoradiotherapy against surgery alone, rather than against perioperative chemotherapy (e.g., MAGIC regimen), which was an emerging standard at the time.
Elderly patients (over 75 years of age) were excluded or underrepresented, limiting the direct generalizability to older or frail populations.

Clinical Significance

The CROSS trial is a landmark study that firmly established neoadjuvant chemoradiotherapy using carboplatin and paclitaxel as a standard of care for locally advanced, resectable esophageal and esophagogastric-junction cancers. By achieving high rates of R0 resection and nearly doubling median overall survival without increasing surgical morbidity or mortality, this well-tolerated regimen revolutionized the multimodality management of upper gastrointestinal malignancies.

Historical Context

Prior to the CROSS trial, the prognosis for locally advanced esophageal cancer managed with surgery alone was poor, plagued by high rates of local recurrence and positive margins. Earlier neoadjuvant trials yielded mixed results and often utilized highly toxic chemotherapy regimens (like cisplatin and fluorouracil) combined with higher-dose radiation, which increased postoperative complications. By introducing a well-tolerated, lower-toxicity regimen with moderate-dose radiation, the CROSS trial demonstrated a profound survival benefit, cementing trimodality therapy as the dominant therapeutic strategy in the Western world.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

The CROSS trial included patients with both squamous cell carcinomas and adenocarcinomas of the esophagus. What are the key risk factors and typical anatomical locations for these two histological subtypes, and what is the basic mechanistic rationale for using chemoradiotherapy before surgery (neoadjuvant) rather than after?

Key Response

This question tests foundational knowledge. Adenocarcinomas typically arise in the distal esophagus/EG junction and are linked to GERD and Barrett's esophagus. Squamous cell carcinomas often occur in the proximal/mid-esophagus and are strongly associated with smoking and alcohol use. Neoadjuvant therapy is mechanistically favored to downstage the tumor, increase the likelihood of achieving negative surgical margins (R0 resection), and eradicate early micrometastatic disease while the local blood supply is intact.

Resident
Resident

In the CROSS trial, the neoadjuvant regimen utilized a backbone of weekly carboplatin and paclitaxel. How does this specific regimen compare in terms of toxicity and tolerability to historical regimens (like cisplatin/5-FU), and which specific patient populations are indicated for this approach?

Key Response

Residents must understand the practical clinical application of this trial. The CROSS regimen (carboplatin/paclitaxel) is significantly better tolerated with lower rates of severe hematologic and gastrointestinal toxicities compared to older cisplatin/5-fluorouracil regimens. It is indicated for patients with locally advanced, resectable (T1N1 or T2-3N0-1) esophageal or EGJ cancer, balancing excellent tumor downstaging with a safety profile that does not significantly increase postoperative mortality.

Fellow
Fellow

The CROSS trial demonstrated a significantly higher pathologic complete response (pCR) rate in squamous cell carcinomas (49%) compared to adenocarcinomas (23%). How does this differential radiosensitivity influence the modern consideration of definitive chemoradiotherapy (organ preservation) for SCC, and what recent phase 3 data dictates adjuvant management for patients who do not achieve a pCR?

Key Response

Fellows must integrate CROSS with modern paradigms. The high pCR rate in SCC has sparked trials evaluating active surveillance (surgery only for salvage) to spare patients esophagectomy morbidity (e.g., SANO trial concepts). Furthermore, fellows should know that based on the CheckMate 577 trial, patients who receive CROSS nCRT followed by surgery but fail to achieve a pCR should be offered adjuvant nivolumab, representing a major paradigm shift in multidisciplinary management.

Attending
Attending

While the CROSS trial established neoadjuvant chemoradiotherapy as a standard of care without increasing 30-day postoperative mortality in a controlled trial setting, how do you navigate the real-world risks of radiation-induced pulmonary complications and anastomotic leaks in older, frailer patients undergoing complex esophagectomies?

Key Response

Attendings must bridge the gap between trial efficacy and real-world effectiveness. Delivering 41.4 Gy of radiation prior to an Ivor Lewis or McKeown esophagectomy can lead to tissue fibrosis and impair anastomotic healing. The attending must thoughtfully weigh oncologic benefit against surgical morbidity, potentially modifying the radiation field, adjusting surgical techniques (e.g., omental flaps), or reconsidering the surgical candidacy of frail patients post-CRT.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

The CROSS trial was powered to detect a survival difference in the overall cohort, combining adenocarcinomas and squamous cell carcinomas despite their distinct biological behaviors and pCR rates. What are the methodological limitations and potential statistical confounding effects of combining these two distinct diseases into a single primary efficacy endpoint, and how would you design a translational study to untangle the drivers of radioresistance in the adenocarcinoma subgroup?

Key Response

This challenges PhDs to critique study design and heterogeneity. Lumping distinct histologies to achieve statistical power can obscure subtype-specific effects; for instance, the immense benefit in SCC might pull the overall survival curve, potentially masking a more modest benefit in adenocarcinoma. A follow-up translational design would require genomic/transcriptomic profiling of pre-treatment adenocarcinoma biopsies to identify biomarkers of radioresistance (e.g., hypoxia signatures, DNA repair mutations) compared to SCC.

Journal Editor
Journal Editor

A critical reviewer examining the CROSS trial might note that the R0 resection rate in the surgery-alone arm was only 69%. How does this relatively low R0 rate in the control group affect the internal validity and the magnitude of the reported survival benefit of the neoadjuvant chemoradiotherapy arm?

Key Response

Editors focus on threats to validity and the performance of the control arm. An R0 rate of 69% in the surgery-alone arm is considered suboptimal by modern, high-volume center standards. A tough reviewer would flag that if the surgical control group underperformed (leaving residual microscopic disease in 31% of patients), the relative benefit of the neoadjuvant arm (which had a 92% R0 rate) might be exaggerated, raising questions about whether better baseline surgical quality could have narrowed the survival gap.

Guideline Committee
Guideline Committee

For adenocarcinomas of the esophagogastric junction (Siewert types I and II), current NCCN and ESMO guidelines support both the CROSS regimen (neoadjuvant chemoradiotherapy) and the FLOT regimen (perioperative chemotherapy). How should a guideline committee grade the evidence and provide recommendations to guide multidisciplinary selection between these two distinct approaches, particularly in light of recent comparative trials like ESOPEC?

Key Response

Guideline committees constantly grapple with overlapping, high-level evidence for competing regimens. Historically, CROSS and FLOT were both Category 1 recommendations without a definitive head-to-head overall survival read-out. A committee must weigh surgical morbidity (radiation effects of CROSS) versus systemic toxicity (FLOT). With recent presentations of trials like ESOPEC showing an overall survival advantage for FLOT in operable adenocarcinomas, guidelines must adapt, potentially shifting CROSS to a preferred option only for SCC or margin-threatened bulky EGJ tumors.

Clinical Landscape

Noteworthy Related Trials

2006

MAGIC Trial

n = 503 · NEJM

Tested

Perioperative ECF chemotherapy

Population

Patients with resectable adenocarcinoma of the stomach, esophagogastric junction, or lower esophagus

Comparator

Surgery alone

Endpoint

Overall survival

Key result: Perioperative chemotherapy significantly improved overall survival compared to surgery alone.
2019

FLOT4-AIO Trial

n = 716 · Lancet

Tested

Perioperative FLOT chemotherapy

Population

Patients with locally advanced, resectable gastric or gastroesophageal junction adenocarcinoma

Comparator

Perioperative ECF or ECX chemotherapy

Endpoint

Overall survival

Key result: FLOT significantly improved overall survival compared to ECF/ECX.
2021

CheckMate 577

n = 794 · NEJM

Tested

Adjuvant nivolumab for 1 year

Population

Patients with resected esophageal or gastroesophageal junction cancer with residual pathologic disease after neoadjuvant chemoradiotherapy

Comparator

Placebo

Endpoint

Disease-free survival

Key result: Adjuvant nivolumab doubled median disease-free survival compared to placebo in patients with residual disease.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis