The New England Journal of Medicine FEBRUARY 11, 2021

Adjuvant Nivolumab in Resected Esophageal or Gastroesophageal Junction Cancer

Ronan J. Kelly, et al. (CheckMate 577 Investigators)

Bottom Line

In patients with resected esophageal or gastroesophageal junction cancer who had residual pathologic disease after neoadjuvant chemoradiotherapy, adjuvant nivolumab significantly prolonged disease-free survival compared with placebo.

Key Findings

1. Nivolumab significantly improved median disease-free survival (DFS) to 22.4 months, compared to 11.0 months with placebo (hazard ratio for disease recurrence or death, 0.69; 95% CI, 0.56 to 0.85; P<0.001).
2. The DFS benefit was consistent across key subgroups, including histological subtype (adenocarcinoma vs. squamous-cell carcinoma) and tumor PD-L1 expression levels.
3. Long-term follow-up (median 78.3 months) sustained the DFS benefit, with a hazard ratio of 0.76 (95% CI, 0.63 to 0.91).
4. While median overall survival showed a numerical improvement with nivolumab (51.7 months) versus placebo (35.3 months), this difference did not reach statistical significance (HR 0.85; P=0.1064).

Study Design

Design
RCT
Double-Blind
Sample
794
Patients
Duration
78.3 mo
Median
Setting
Multicenter, global
Population Adults with resected (R0) stage II/III esophageal or gastroesophageal junction cancer who had residual pathologic disease after neoadjuvant chemoradiotherapy
Intervention Nivolumab 240 mg intravenously every 2 weeks for 16 weeks, followed by 480 mg every 4 weeks
Comparator Placebo administered on a matching schedule
Outcome Disease-free survival (DFS)

Study Limitations

The trial did not demonstrate a statistically significant improvement in overall survival, a key secondary endpoint.
The study focused exclusively on patients with residual pathologic disease, limiting generalizability to those who achieve a pathological complete response after neoadjuvant therapy.
Potential for long-term immune-related adverse events requires continued monitoring.

Clinical Significance

CheckMate-577 established adjuvant nivolumab as a new standard of care for patients with resected esophageal or gastroesophageal junction cancer who have residual disease following neoadjuvant chemoradiotherapy, addressing a significant unmet need in this high-risk population.

Historical Context

Prior to this trial, patients who underwent neoadjuvant chemoradiotherapy followed by surgical resection for esophageal or gastroesophageal junction cancer but failed to achieve a pathological complete response lacked effective adjuvant systemic therapies and faced high rates of disease recurrence.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

What is the physiological rationale for using a PD-1 inhibitor specifically in patients who have residual pathologic disease after neoadjuvant chemoradiotherapy (nCRT), rather than those who achieve a pathologic complete response?

Key Response

Residual disease after nCRT is an indicator of resistance to initial therapy and a high risk for recurrence. Neoadjuvant radiation can induce 'immunogenic cell death' and increase PD-L1 expression and T-cell infiltration. Nivolumab (a PD-1 inhibitor) leverages this primed immune environment to eliminate microscopic residual disease that the initial treatment failed to eradicate, whereas patients with a complete response already have a significantly lower risk profile where the benefit-to-risk ratio of immunotherapy is less clear.

Resident
Resident

A patient with esophageal adenocarcinoma undergoes the CROSS regimen (carboplatin/paclitaxel/radiation) followed by surgery, and the pathology shows a Mandard grade 3 response (residual cancer cells). According to CheckMate-577, what is the recommended management and what specific benefit should be discussed with the patient?

Key Response

The patient should be offered adjuvant nivolumab for up to one year. According to the study, nivolumab doubled the median disease-free survival (DFS) compared to placebo (22.4 months vs. 11.0 months; HR 0.69). It is critical to note that this recommendation applies only to those with residual pathologic disease (non-pCR) following neoadjuvant chemoradiotherapy, not those who underwent surgery first or received perioperative chemotherapy alone.

Fellow
Fellow

While CheckMate-577 showed benefit across subtypes, the hazard ratio for disease-free survival was more pronounced in Squamous Cell Carcinoma (SCC) than in Adenocarcinoma (AC). How should this subgroup analysis and the difference in PD-L1 expression patterns influence your clinical expectation of therapy efficacy?

Key Response

In the study, the HR for SCC was 0.61 compared to 0.75 for AC. This aligns with a broader trend in oncology where SCCs often show higher mutational burdens and greater responsiveness to PD-1 blockade. While the benefit is statistically significant for both, the fellow should recognize that GEJ adenocarcinoma patients—especially those with low CPS scores—might derive less absolute benefit than those with esophageal SCC, necessitating nuanced counseling regarding the 12-month treatment duration.

Attending
Attending

CheckMate-577 established a new standard of care based on Disease-Free Survival (DFS). In the absence of mature Overall Survival (OS) data at the time of the primary analysis, how do you justify the use of a potentially toxic and expensive year-long therapy to a patient who is already recovering from a highly morbid esophagectomy?

Key Response

The justification lies in the doubling of median DFS and the high recurrence rate in this population. Recurrent esophageal cancer is almost universally fatal and significantly impacts quality of life. By delaying or preventing recurrence, nivolumab provides a clear clinical benefit. Furthermore, the safety profile in the study showed that grade 3-4 immune-related adverse events were relatively low (13%), and the majority of patients were able to maintain a high quality of life during the adjuvant phase.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

Critique the use of 'pathologic residual disease' as both a predictive biomarker and a requirement for study inclusion. How does this selection bias affect the calculation of the treatment's effect size compared to an unselected intention-to-treat population?

Key Response

By enrolling only non-pCR patients, the study enriches for a 'high-risk' population that is biologically predisposed to fail traditional cytotoxic therapy. While this increases the power to detect a DFS difference (as events occur faster), it limits the generalizability of the results to pCR patients. From a research design perspective, this is an 'enrichment strategy' that may overstate the relative benefit if one were to apply it to all-comers, but it provides a high-level of evidence for the specific subset with the greatest unmet clinical need.

Journal Editor
Journal Editor

A major criticism of CheckMate-577 is the lack of a standardized assessment for 'residual disease' across international sites and the potential for 'crossover' to subsequent immunotherapy upon recurrence. How might these factors compromise the internal and external validity of the DFS and OS endpoints?

Key Response

Variation in pathologic reporting (e.g., different systems for grading regression) can introduce noise into the inclusion criteria, though the binary presence/absence of cancer cells is generally robust. The more significant issue for an editor is the OS endpoint: if placebo-arm patients receive PD-1 inhibitors upon recurrence (crossover), the OS benefit of adjuvant therapy may be masked. A reviewer would demand a breakdown of post-progression therapies to determine if the adjuvant timing is truly superior to salvage immunotherapy.

Guideline Committee
Guideline Committee

The NCCN and ESMO guidelines now list adjuvant nivolumab as a Category 1 recommendation for resected esophageal/GEJ cancer with residual disease. How should the committee address the application of this evidence to patients who received perioperative chemotherapy (like the FLOT regimen) rather than the nCRT required in CheckMate-577?

Key Response

Current guidelines (e.g., NCCN) strictly specify that nivolumab is indicated after chemoradiotherapy (the CROSS or similar regimens). CheckMate-577 specifically tested the synergy between radiation-induced immunogenicity and PD-1 inhibition. There is currently no Category 1 evidence that nivolumab provides the same benefit after perioperative chemotherapy alone (FLOT). The committee must decide whether to allow 'extrapolation' or to maintain a strict 'evidence-based' boundary, usually opting for the latter until trials like CheckMate-58T or similar perioperative IO trials provide specific data.

Clinical Landscape

Noteworthy Related Trials

2012

CROSS Trial

n = 368 · NEJM

Tested

Neoadjuvant chemoradiotherapy with carboplatin and paclitaxel followed by surgery

Population

Resectable esophageal or gastroesophageal junction cancer

Comparator

Surgery alone

Endpoint

Overall survival

Key result: Neoadjuvant chemoradiotherapy significantly improved median overall survival compared to surgery alone.
2021

CheckMate-649

n = 1581 · NEJM

Tested

Nivolumab plus chemotherapy

Population

Advanced or metastatic gastric, gastroesophageal junction, or esophageal adenocarcinoma

Comparator

Chemotherapy alone

Endpoint

Overall survival and progression-free survival in PD-L1 positive patients

Key result: Nivolumab plus chemotherapy demonstrated superior overall and progression-free survival in patients with advanced disease.
2023

TOPGEAR Trial

n = 574 · Lancet Oncol

Tested

Perioperative chemoradiotherapy

Population

Resectable gastric or gastroesophageal junction adenocarcinoma

Comparator

Perioperative chemotherapy alone

Endpoint

Overall survival

Key result: The addition of preoperative chemoradiotherapy to perioperative chemotherapy did not show an improvement in overall survival.

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