Neoadjuvant cemiplimab leads to tumor necrosis in HCC
Click Here to Manage Email Alerts
More than one-third of a small cohort patients with early-stage hepatocellular carcinoma achieved at least 50% tumor necrosis following neoadjuvant treatment with cemiplimab, results of a phase 2a trial showed.
The anti-PD-1 antibody also had an acceptable safety profile, according to the findings, presented during the virtual American Association for Cancer Research Annual Meeting.
“In many different cancer types, we’ve seen that neoadjuvant immunotherapy has significant benefit as far as a pathologic response, shrinking the tumors,” Thomas U. Marron, MD, PhD, assistant director of immunotherapy and early phase trials at Tisch Cancer Institute and assistant professor of medicine, hematology and medical oncology at Icahn School of Medicine at Mount Sinai, told Healio. “The overall goal of these trials is not necessarily the pathologic response, which is our primary endpoint — that’s more of a surrogate for a systemic immune response that’s going to impart protection. The overall goal is to decrease the likelihood that these cancers come back.”
Surgical resection is a standard first-line treatment for early-stage HCC, even though the majority of tumors recur. Because recurrence is believed to be the result of the micrometastases that persist after resection, researchers sought to take advantage of the potential benefit of neoadjuvant therapy in improving HCC outcomes.
The open-label, multicenter trial enrolled 21 adults (median age, 68 years; range, 45-82; 85.7% men; 52.4% Asian) with resectable HCC over the course of 18 months. Patients received two cycles of 350 mg cemiplimab (Libtayo; Regeneron Pharmaceuticals, Sanofi) via IV every 3 weeks, followed by surgical resection and an additional eight adjuvant cycles.
Significant tumor necrosis, defined as more than 70% necrosis of the resected tumor, served as the study’s primary endpoint.
One patient was found to have metastatic disease at the time of surgery and resection was aborted.
Among the 20 patients with resected tumors, four (20%) met the predefined endpoint of more than 70% tumor necrosis and seven (35%) had at least 50% tumor necrosis. Additionally, three (15%) of the four patients with more than 70% tumor necrosis had a pathologic complete response (100% tumor necrosis).
“We were excited to see that 35% of the patients had a lot of tumor necrosis,” Marron said. “But in the end, of the outcomes that we’re most interested in, one obviously is recurrence-free survival. What’s very exciting is the analysis that we’re doing right now at the single-cell level and the histologic level in identifying which subset of patients successfully activated an immune response. That hopefully will have some nice immunologic surrogates that will tell us whether a patient responds or not.”
Results of initial pathologic assessment suggested a correlation of immune cell infiltration and tumor necrosis.
Most patients (90.5%) experienced at least one treatment-emergent adverse event, the most common of which included increased aspartate aminotransferase (any grade, 28.6%; grade 3, 4.8%), increased blood creatine phosphokinase (any grade, 14.3%; grade 3, 9.5%) and fatigue (any grade, 14.3%; grade 3, 0%).
Marron said researchers need a much larger patient population “to really have a piece of data we can hang our hat on” and could take the research further by testing different combinations of drugs as neoadjuvant therapy for patients with HCC.
“We’re looking at combinations of a PD-1-blocking antibody and agents for which we already have a lot of safety data that we would to a patient for about a month after surgery to try to leach out all of those immunosuppressive immune cells, so that the activating T cells can actually go to work and kill the tumor,” he said.
Perspective
Back to Top
Ghassan K. Abou-Alfa, MD, MBA
Marron and colleagues sought to determine whether checkpoint inhibitors can generate a response from a neoadjuvant perspective.
Their study has a very limited amount of data, with only 21 patients. At the same time, it is intriguing that the researchers did not specify whether the enrolled patients already had been determined to have resectable disease. It’s a critical component because it can help test the tumor response to systemic therapy to make sure that it’s benefiting from that therapy before surgery, when generally patients would tolerate such therapy better.
If you give neoadjuvant therapy to a patient with a tumor that’s already resectable to ensure the chance of recurrence is way lower, that’s a very legitimate component to the work. On the other hand, if you have a tumor that is not resectable and try to make it resectable, what we call conversion therapy, that’s not yet an amenable approach in liver cancer.
Think about it logically: Say you have a tumor that is 10 cm and, for whatever reason, the surgeon can’t take it out, so you shrink it to 5 cm and then take it out. Now, let’s assume there’s another patient with a 5 cm tumor to begin with. Can you tell that the patient with the 5 cm tumor to begin with has a tumor of the same biology as the one with a 10 cm tumor that was reduced to 5 cm? That the tumor grew to 10 cm implied a biological event that caused it to grow.
In such an instance, the researchers need to justify and explain whether the biology of the tumor changed because of the checkpoint inhibitor, to have a more successful surgery despite the fact that it wasn’t resectable to begin with.
The abstract presentation implies all the patients had resectable tumors to begin with and were given neoadjuvant therapy. If so, it is great to see researchers found necrosis; however, this occurs in many other instances. But this brings up a very important component: How can you prove, in a small study like this, that ultimately this presumed neoadjuvant therapy intervention would improve outcomes?
Collapse
Marron TU, et al. Abstract CT182. Presented at: American Association for Cancer Research Annual Meeting (virtual meeting); April 10-15, 2021.
Click Here to Manage Email Alerts