Preoperative pembrolizumab may enhance major pathologic response in esophageal cancer
Click Here to Manage Email Alerts
Adding pembrolizumab to preoperative chemoradiotherapy increased major pathologic response rate compared with historical data among patients with esophageal adenocarcinoma, according to a study presented at the virtual ASCO Annual Meeting.
Additionally, findings from the multicenter, randomized phase 2 study showed use of pembrolizumab (Keytruda, Merck) in the preoperative setting was safe, with the higher rates of major pathologic response correlating with improved survival and occurring more frequently among patients with esophageal adenocarcinoma than those with gastroesophageal junction tumors.
“Pembrolizumab specifically, and checkpoint inhibition in general, has been examined in upper gastrointestinal cancers for some time. We knew that there was modest activity in esophageal cancer and wanted to examine the potential benefit of chemotherapy and chemoradiation in improving the efficacy of checkpoint inhibition,” Manish A. Shah, MD, director of the gastrointestinal oncology program at Weill Cornell Medicine and chief of the solid tumor service and co-director of Center for Advanced Digestive Disease at NewYork-Presbyterian, told Healio.
“Esophageal cancer is a challenging disease, and the majority of patients who are able to undergo preoperative therapy and resection still have recurrence,” Shah added. “We wanted to investigate if the addition of preoperative pembrolizumab could improve the efficacy of standard-of-care preoperative treatment.”
The study included 39 patients (median age, 67.6 years; range, 38.4-81.1; 79.5% men; 92.3% white) with clinical stage T2N+ or T3-4 Nany, M0 esophageal adenocarcinoma eligible for curative surgery. Fifteen patients had esophagus/gastroesophageal junction type 1 disease, and the other 24 had type II or type III.
Researchers randomly assigned patients 1:1 to receive full-dose paclitaxel-carboplatin (n = 20) or paclitaxel-carboplatin plus pembrolizumab (n = 19) induction therapy. All patients then received chemoradiotherapy with weekly paclitaxel-carboplatin, radiotherapy (41.4 Gy in 23 fractions), and pembrolizumab every 3 weeks, followed by pembrolizumab for 1 year after resection.
The rate of major pathologic response, defined as pathologic complete response or near-complete response with less than 10% residual cancer, and safety served as the study’s primary endpoints. The study was designed to improve the major pathologic response rate from the 30% historical rate to 47% with the addition of pembrolizumab, with 80% power and 0.1 one-sided significance level among 36 patients.
Secondary endpoints included median and 1-year DFS and OS and, in an ongoing analysis, researchers also seek to evaluate changes in the tumor microenvironment with and without pembrolizumab based on the randomization in the induction setting.
Researchers reported 1-year DFS of 60.4% (95% CI, 39.3-76.2) and 1-year OS of 77.5% (95% CI, 56.4-89.3).
The major pathologic response rate was 48.7% (95% CI, 33-64.4), meeting the study’s primary endpoint.
Researchers reported a 1-year DFS of 100% among patients with a major pathologic response vs. 23.5% among those without such a response (P = .001). A greater proportion of patients with major pathologic response also achieved 1-year OS (93.8% vs. 62.5%; P = .04).
“The addition of preoperative immunotherapy to chemoradiotherapy seems to increase the major pathologic response rate to therapy,” Shah said. “We demonstrated that a major pathologic response is significantly associated with better survival, both DFS and OS, than patients who did not have a major pathologic response. This may have important ramifications for further drug development in esophageal cancer.”
Researchers also conducted single-cell RNA sequencing on over 100,000 single cells collected from baseline endoscopic biopsies of 33 patients and found what Shah described as a “complex immune microenvironment.” Patients with higher proportion of activated B cells and plasmacytoid dendric cells in their tumors at baseline appeared less likely to achieve pathologic response, whereas enrichment in monocytes and cytotoxic CD8+ T cells increased likelihood of pathologic response, according to Shah.
“We were surprised to find potentially important differences in the tumor microenvironment in patients who had a major pathologic response and in those without a major pathologic response,” he told Healio.
Researchers also observed a significantly higher major pathologic response rate among patients with esophageal cancer compared with gastroesophageal junction tumors (73.3% vs. 33.3%; P = .02). These data correlated with the findings of the tumor microenvironment analysis, as active B cells and plasmacytoid dendritic cells appeared to be enriched in gastroesophageal junction tumors.
Results showed chemoradiotherapy was well-tolerated. Common adverse events in the neoadjuvant setting including neutropenia (grade 2, 20.5%; grades 3-4, 30.8%), fatigue/weakness (grade 2, 38.5%; grades 3-4, 33.3%) and nausea/vomiting/dehydration (grade 2, 41%; grades 3-4, 12.8%). Minimal events were related to immunotherapy; three patients experienced rash or erythema multiforme and one patient had pneumonitis.
Postoperative events also appeared typical, Shah said, and including wound dehiscence (grade 2, 2.8%; grades 3-4, 5.6%), infection (grade 2, 5.6%; grades 3-4, 11.1%) and atrial fibrillation/arrythmia (grade 2, 2.8%; grades 3-4, 13.9%). Grade 3 to grade 4 toxicity related to immunotherapy in the postoperative setting included elevated liver enzymes (13.9%), elevated blood sugar (8.3%), pneumonitis (11.1%) and adrenal insufficiency (2.8%).
Shah said researchers need to better understand the role of the immune microenvironment in response to chemotherapy and immunotherapy.
“This may have important broadly reaching applications,” he said. “We also need to work toward improving the major pathologic response rate, which may be an early indicator of the potential benefits of a particular therapeutic strategy.”
Perspective
Back to Top
Immune checkpoint inhibitors are transforming the way we treat patients with esophageal and gastroesophageal junction adenocarcinomas. Recent phase 3 data from KEYNOTE-590 and CheckMate 649 support the incorporation of anti-PD-1 monoclonal antibody therapy into front-line systemic treatment regimens for patients with metastatic esophageal or gastroesophageal junction cancers, with the most significant impact on clinical outcomes for patients whose tumors show high PD-L1 expression. Additionally, single-agent nivolumab (Opdivo, Bristol Myers Squibb) has emerged as a new standard of care in the postoperative adjuvant setting for resected patients who do not achieve a pathologic complete response following neoadjuvant chemoradiation based on CheckMate 577.
On these bases, there is considerable and understandable interest in evaluating the role immunotherapy may play within other clinical contexts for esophageal/ gastroesophageal junction cancer, such as in the preoperative setting. Shah and colleagues present promising results from a randomized, multicenter phase 2 trial, in which patients with resectable disease were randomly assigned to receive induction chemotherapy plus chemoradiation prior to surgery, using a standard carboplatin/paclitaxel platform; or this same sequential chemotherapy/chemoradiation combination with the addition of concurrent pembrolizumab.
It is of highest priority in this potentially curative setting that we feel confident that the integration of immunotherapy within a standard neoadjuvant treatment framework does not compromise patients’ ability to successfully undergo subsequent surgery, that such treatment won’t produce significant delays in getting to surgery, and that the regimen is not associated with an unacceptably high rate of peri-/postoperative complications. Deciding what meets the threshold of “safe” can be challenging, given that adverse events associated with both neoadjuvant treatment and a major cancer operation, as well as the underlying disease itself, are commonplace in this patient population during traditional trimodality therapy. With these caveats in mind, Shah and colleagues quite reasonably state the study regimen appears sufficiently safe; the toxicities observed on their trial are neither unexpected nor deal-breakers in terms of moving forward with further investigation.
Shah and colleagues also choose to cast a slightly broader net in terms of defining clinical efficacy in this study, subsuming within their primary endpoint not just pathologic complete responses, but also “near-complete” responses (defined as < 10% residual cancer identified in the pathologic specimen). This major pathologic response rate of 48.7% is indeed promising and appears to be associated with improved DFS in their study cohort, recognizing the need for both longer follow-up and further validation in larger numbers of patients.
Overall, this study represents an important proof of concept: namely, the safety, feasibility and preliminary evidence of efficacy using a combined immunotherapy-plus-chemotherapy/chemoradiation approach in the neoadjuvant setting for esophageal/ gastroesophageal junction cancers. Although it may already be tempting to introduce anti-PD-1 inhibitors into our current treatment paradigms for this specific context, especially for patients whose tumors we know have very elevated levels of PD-L1 expression, ideally, we should await much further data before adopting this into routine clinical practice.
References:
The following were presented at European Society for Medical Oncology Virtual Congress 2020; Sept. 19-21, 2020.
Kato K, et al. Abstract LBA8_PR.
Kelly RJ, et al. Abstract LBA9_PR.
Moehler M, et al. Abstract LBA6_PR.
University of California, San Francisco
Collapse
Shah MA, et al. Abstract 4005. Presented at: ASCO Annual Meeting (virtual meeting); June 4-8, 2021.
Click Here to Manage Email Alerts