SAN FRANCISCO — Almost 80% of patients with locally advanced esophagogastric adenocarcinoma remained alive without relapse at 1 year with adjuvant durvalumab (Imfinzi) for residual disease after initial trimodal therapy, a small prospective trial showed.
The 24-patient study population had a 12-month relapse-free survival (RFS) rate of 79.2% and an estimated 67.9% probability of survival without relapse at 26 months. Though not sufficient to call for a change in clinical practice, the findings set the stage for a future randomized phase III trial with the potential to establish adjuvant durvalumab as a treatment standard for patients with residual disease, as reported here at the Gastrointestinal Cancers Symposium.
“These results are based on a very small, nonrandomized study, so I wouldn’t necessarily say that this is practice changing,” said Hirva Mamdani, MD, of Karmanos Cancer Institute and Wayne State University in Detroit, who reported the data. “It is definitely encouraging. We have not seen the [survival] curves flattening out like this in esophageal cancer for a very long time, but we need to wait for randomized data.”
The standard approach to locally advanced esophageal and gastroesophageal junction (GEJ) adenocarcinoma consists of concurrent chemoradiation followed by esophagectomy. About half of patients have disease relapse within a year after completing treatment.
Patients who do not achieve a pathologic complete response and have persistent nodal disease with standard treatment have a heightened risk of relapse, said Mamdani. To date, no adjuvant therapy provided a survival benefit in this setting.
The rationale for evaluating adjuvant treatment with the PD-L1 inhibitor durvalumab included evidence of anti-PD-1/PD-L1 activity in metastatic PD-L1-positive esophageal cancer and preclinical evidence that radiation therapy and chemotherapy induce upregulation of the PD-1/PD-L1 pathway.
Mamdani reported findings from a multicenter phase II trial involving patients with locally advanced esophageal/GEJ adenocarcinoma treated with chemoradiation and gastrectomy that led to disease-free surgical margins (R0). Patients were unselected with regard to tumor PD-L1 expression status.
The trial had a primary endpoint of 1-year RFS, and secondary endpoints included safety and feasibility of administering adjuvant durvalumab. Investigators hypothesized that adjuvant durvalumab would increase 1-year RFS by 50%, from 50% in a historical control group treated with standard therapy to 75%.
All patients received platinum-containing chemotherapy (carboplatin/paclitaxel or cisplatin/5FU) and definitive radiation therapy (total dose 41 Gy). Within 1 to 3 months after gastrectomy, patients received durvalumab every 4 weeks for a maximum of 12 months. Follow-up for the primary endpoint continued for a year after treatment discontinuation. Disease assessment at 3-month intervals included CT of the chest, abdomen, and pelvis.
The study cohort had a median age of 60, and men accounted for all but one of the patients. The primary disease site was the GEJ in 14 patients and distal esophagus in 10. Nodal status was N0 in five patients, N1 in seven, N2 in nine, and N3 in three.
The patients received a median of 12.5 cycles of durvalumab, and half the patients completed the planned 12 months of adjuvant therapy. Six patients discontinued treatment because of progressive disease, five because of adverse events, and one withdrew consent.
The most commonly reported grade 1/2 adverse events were fatigue (33%) and nausea (25%). Three patients discontinued treatment because of treatment-related grade 3 adverse events (one case each of pneumonitis, hepatitis, and colitis). Additionally, one patient each developed hypoglycemia and hyperglycemia unrelated to study treatment. No grade ≥4 adverse events occurred during the study.
After a median follow-up of 14.5 months, the trial met the primary endpoint of 1-year RFS ≥75%.
The presentation sparked numerous questions from the audience, beginning with the decision not to limit enrollment to patients with PD-L1-positive tumors, which presumably would have improved the response potential with durvalumab.
“The thought process behind including all comers was that, in the post-chemoradiation setting, we do not know what the biomarker of response to immunotherapy would be,” said Mamdani. “It will be very useful to have all comers and analyze the data.”
Investigators will perform several correlational analyses, including studies of the relationship of RFS with PD-L1 status and with tumor infiltrating lymphocytes.
In response to a question from the audience, Mamdani acknowledged an ongoing randomized, placebo-controlled phase III trial of adjuvant nivolumab (Opdivo) in patients with resected esophageal/GEJ cancer. She said the trial will provide additional information about the potential role of adjuvant immune checkpoint inhibition, including “how PD-L1 status and some of the other immune markers play out.”
Collaborators in the investigator-initiated trial included Medimmune, AstraZeneca, and the Big Ten Cancer Research Consortium.
Mamdani disclosed a relationship with Tempus.