Ronan J. Kelly, MD, MBA, FASCO
The long-term disease-free survival (DFS) benefit and numerical overall survival improvement seen with nivolumab (Opdivo) vs placebo in patients with resected esophageal or gastroesophageal junction (GEJ) cancer who have residual pathologic disease following neoadjuvant chemoradiotherapy reinforces the role of adjuvant nivolumab as a standard of care (SOC) in this patient population, according to Ronan J. Kelly, MD, MBA, FASCO.
Findings from the final OS and updated DFS data from the phase 3 CheckMate 577 study (NCT02743494) were presented during the 2025 ASCO Annual Meeting. With a minimum follow-up of 5 years, the median DFS increased from 10.8 months (95% CI, 8.3-14.3) with placebo (n = 262) to 21.8 months (95% CI, 16.6-29.7) with adjuvant nivolumab (n = 532; HR, 0.76; 95% CI, 0.63-0.91). The median OS was also prolonged by 16.4 months with nivolumab vs placebo; however, this difference did not meet the threshold for statistical significance (HR, 0.85; 95.87% CI, 0.70-1.04; P = .1064). The 5-year OS rate was also higher in the nivolumab vs placebo arm. Regarding safety, nivolumab continued to demonstrate a favorable toxicity profile with no new safety signals observed over extended follow-up.
“[The data] suggest a clinically meaningful improvement in OS [with nivolumab], although statistical significance was not met. Looking at the OS subgroup analysis, however, did show that the majority of groups favored nivolumab,” Kelly, the chief of oncology, the W.W. Caruth, Jr. Chair in Immunology, and the founder and director of the Texas Cancer Interception Institute at Baylor Scott & White Health in Dallas, shared with OncLive®.
In the interview, Kelly provided an in-depth look at the final efficacy and safety outcomes from CheckMate 577, highlighted the role of PD-L1 expression and histology in response, and emphasized evolving treatment paradigms, with perioperative regimens gaining favor for GEJ and gastric cancers while adjuvant nivolumab remains standard for esophageal squamous cell carcinoma.
OncLive: What were the primary findings previously reported from CheckMate 577?
Kelly: CheckMate 577 was originally presented at the 2020 ESMO Congress, where a statistically significant and clinically meaningful DFS benefit [was demonstrated] with adjuvant nivolumab compared with placebo. The study also showed a well-tolerated safety profile in patients with resected esophageal or GEJ cancers following trimodality therapy.
Based on these findings, which were published in The New England Journal of Medicine in 2021, nivolumab was globally approved for the adjuvant treatment of patients with resected esophageal or GEJ cancer and residual pathologic disease following neoadjuvant chemoradiation. To date, adjuvant nivolumab remains the only approved immuno-oncology agent in operable gastroesophageal cancer. However, OS data have remained an unmet need in this setting.
What was the design of CheckMate 577?
The final analysis of the secondary end point of OS was presented from the CheckMate 577 study, along with 5-year follow-up data for DFS and exploratory end points. This global, phase 3, randomized, double-blind, placebo-controlled trial enrolled patients with stage II or III esophageal or GEJ cancer, including both adenocarcinoma and squamous cell carcinoma histologies. All patients were required to have received neoadjuvant chemoradiation followed by surgical resection.
Importantly, patients with a pathologic complete response, [which includes] approximately 25% to 30% of this population, were excluded. The study focused on patients with residual pathologic disease, indicative of more aggressive tumor biology and a lack of response to neoadjuvant therapy. A total of 794 patients were [randomly assigned] in a 2:1 ratio to receive either adjuvant nivolumab or placebo for a total duration of 1 year.
The primary end point was DFS. Secondary end points included OS rates at 1, 2, and 3 years, with exploratory analyses evaluating safety and distant metastasis-free survival. [Notably,] there was an approximate 6-month interval between initiation of chemoradiation and randomization, reflecting the time required for chemoradiation, surgery, and recovery prior to trial enrollment.
What should be known about the baseline characteristics of patients in this study?
In terms of the baseline characteristics for patients enrolled in the study [who received nivolumab], the salient points to highlight [are as follows]: tumor location was esophageal in 59% of patients and gastroesophageal junction in 41%. Adenocarcinoma was the predominant histology, occurring in 71% of patients, while 29% had squamous cell carcinoma. The poor prognostic factor of lymph node–positive disease, specifically ypN1, was present in 58% of patients.
PD-L1 expression [was assessed] using both tumor proportion score (TPS) and combined positive score [CPS]. We know that in this disease setting, CPS is a better assessment of PD-L1 status. For the first time, this analysis showed that approximately 10% of patients had immunologically cold tumors, defined as CPS less than 1. The remaining patients had CPS of 1 or greater, indicating immunologically hotter tumors.
Regarding treatment exposure and discontinuation, no patients [remained] on treatment with the study drug [at the time of this analysis]. [A total of] 49% completed treatment. Among those who discontinued treatment, 29% [did so due to] disease progression in the nivolumab arm, compared with 44% in the placebo arm. The median duration of treatment and the median number of doses were similar between arms. We did see more deaths due to disease progression in the placebo arm compared with the nivolumab arm. There were no on-study treatment-related deaths in the nivolumab arm.
What efficacy and safety data were presented during the 2025 ASCO Annual Meeting?
At the 5-year follow-up, [adjuvant nivolumab] continued to demonstrate a clinically meaningful improvement [in DFS] vs placebo. With this longer follow-up, the median disease-free survival (DFS) was 21.8 months with nivolumab and 10.8 months with placebo, with a HR of 0.76, maintaining the doubling in DFS initially observed. [The Kaplan-Meier curves] separated at approximately 3 to 4 months and remained separated. The 5-year DFS rate was 37% in the nivolumab arm and 29% in the placebo group.
Distant metastasis–free survival [DMFS] was an exploratory end point. Again, we showed that nivolumab prevents distant metastasis. The median [DMFS] was 27.3 months vs 14.6 months for placebo, with a HR of 0.75. The Kaplan-Meier curve separated at the 4-month mark and remained separated. The 5-year DMFS rate was 38% in the nivolumab arm and 29% in the placebo arm.
OS was a secondary end point. For the first time, we showed the final OS results after a minimum follow-up of 5 years. The median OS was 16.4 months longer with nivolumab vs placebo, with a median OS of 35.3 months in the placebo arm vs 51.7 months with nivolumab. The 5-year OS rate was 46% with nivolumab and 41% with placebo, indicating a clinically meaningful improvement, although the difference did not reach statistical significance with a HR of 0.85.
In terms of safety, nivolumab was well tolerated, and the incidence of treatment related adverse effects remains consistent with previous reported results.
What did subgroup analyses reveal about the benefit of adjuvant nivolumab in select patient subgroups, as well as the effect of subsequent systemic therapy on outcomes?
If we look at patients with esophageal cancer, [they derived benefit], with an HR of 0.69. However, we saw limited to no efficacy in patients with GEJ cancer, where the HR was 1.14.
[Patients with] both histologies—adenocarcinoma and squamous cell carcinoma—responded, with a HR of 0.72 for squamous histology. [When stratified by] PD-L1 combined positive score [CPS], patients with a CPS of 1 or greater, [defined as] immunologically “hot,” had improved outcomes, with an HR of 0.79. [Conversely,] those with a CPS less than 1, [defined as] immunologically “cold,” had a HR of 1.4, suggesting limited benefit in this subgroup. [However,] this CPS less than 1 group [represented] only approximately 10% of patients. We should have some degree of caution in interpreting these data, because the subgroups are small and were exploratory in nature. The study also wasn’t powered to look at this, although it mirrors what we’ve seen in the metastatic setting, where those patients with cold tumors are not deriving much benefit from immune checkpoint inhibition in esophageal and gastric cancer.
Finally, given that patients were treated with 1 year of nivolumab and followed for a minimum of 5 years, the effect of subsequent systemic therapy was evaluated. In the placebo arm, approximately 60% of patients received subsequent treatment compared with 46% in the nivolumab arm. If we break that down into systemic treatment, which is the most important in this disease setting when patients progress, approximately 50% of placebo-treated patients received systemic therapy vs 37% in the nivolumab arm. This makes it challenging, because now you have an imbalance between the treatment arms moving forward. To account for the impact of the confounding effects of subsequent treatment, a 2-stage adjustment method was used. When we were able to account for the effect of subsequent treatments, and show the effect of nivolumab alone, the median OS was 38.6 months vs 20.2 months with placebo, with an adjusted HR of 0.73. The Kaplan-Meier curves separated early and remained separated, indicating that subsequent treatment does have an effect in terms of trying to interpret OS.
[Overall, adjuvant nivolumab] was well tolerated, and we think these results further support adjuvant as a standard of care in this disease setting. [It’s interesting [to note] that the FDA has approved subcutaneous nivolumab for patients with resected esophageal and GEJ cancer with residual disease after trimodality therapy, so that offers an alternative route for administration.
What do these findings suggest about the role of adjuvant nivolumab and other perioperative immunotherapy regimens more broadly in esophageal and GEJ cancers?
There has been a clear decline in the use of radiation therapy for GEJ and gastric cancers. The current trajectory suggests that these tumor types should not be treated with neoadjuvant chemoradiation. Evidence strongly supports perioperative treatment as the more appropriate strategy in this setting. With durvalumab [Imfinzi] demonstrating benefit over FLOT alone in the phase 3 MATTERHORN [NCT04592913] study, this regimen is anticipated to become a new SOC for GEJ and gastric cancers.
In contrast, for patients with esophageal squamous cell carcinoma, radiation continues to play an important role. The treatment approach utilized in the CheckMate 577 study, involving neoadjuvant chemoradiation followed by surgery and 1 year of adjuvant nivolumab, remains the SOC. Notably, among patients with esophageal squamous cell carcinoma, the trial demonstrated a 19-month improvement in OS with nivolumab, underscoring its efficacy in this subgroup.
For patients with esophageal adenocarcinoma—a group not included in the MATTERHORN study—prior data from the [phase 3] German ESOPEC trial [NCT02509286] showed a benefit for perioperative chemotherapy compared with neoadjuvant chemoradiation. However, this trial did not compare perioperative chemotherapy with chemoradiation followed by adjuvant nivolumab, which is currently the standard in many practices.
As such, treatment decisions in this population must be individualized. For younger, fit patients who can tolerate FLOT, perioperative chemotherapy is expected to become standard. However, older or more frail patients in community settings may not tolerate this regimen well. There is variability in the ability to administer FLOT safely, and treatment decisions must consider the patient’s fitness to proceed to surgery. Avoiding overtreatment that could preclude surgical resection is a critical consideration.
In these cases, oncologists may choose to pursue neoadjuvant chemoradiation followed by adjuvant nivolumab, based on the CheckMate 577 data. Ultimately, the treating physician is best positioned to assess patient suitability for intensive therapy. Whether durvalumab will eventually be incorporated into the treatment of esophageal adenocarcinoma remains to be determined, pending FDA guidance and updates to National Comprehensive Cancer Network Guidelines, as MATTERHORN did not include this population. For now, the field appears to be coalescing around perioperative chemotherapy for gastric and GEJ tumors, chemoradiation followed by adjuvant nivolumab for squamous histology, and individualized strategies for esophageal adenocarcinoma based on fitness for treatment.
Reference
Kelly RJ, Ajani JA, Kuzdzal J, et al. Adjuvant nivolumab in resected esophageal or gastroesophageal junction cancer (EC/GEJC) following neoadjuvant chemoradiotherapy (CRT): First results of overall survival (OS) from CheckMate 577. J Clin Oncol. 2025;43(suppl 16):4000. doi:10.1200/JCO.2025.43.16_suppl.4000