Neoantigen-specific tumor-infiltrating lymphocytes (TILs) plus pembrolizumab demonstrate efficacy against treatment-refractory gastrointestinal (GI) cancers in early results from a phase 2 trial.
This is the first clinical trial to show that TILs can deliver responses in patients with metastatic GI cancers, making it a notable advance in the efforts to adapt cellular immunotherapy to solid tumors beyond melanoma, lead author Frank J. Lowery, PhD, of the National Cancer Institute, Bethesda, Maryland, and colleagues reported.
How Did This Trial Differ From Previous Studies Involving TILs for GI Cancers?
In early 2024, the US Food and Drug Administration (FDA) approved the first and currently only commercially available TIL therapy, lifileucel, for advanced melanoma. Similar approaches have failed to show efficacy in GI cancers, likely due to a high proportion of nonreactive or exhausted T cells in the infusion product, the investigators wrote in Nature Medicine.
To address this limitation, Lowery and colleagues developed a strategy to select and expand only those TIL cultures that exhibited reactivity to patient-specific neoantigens — mutations uniquely expressed in each patient’s tumor.
This approach was informed by a prior case in which a patient with pancreatic cancer who had failed unselected TIL therapy experienced a durable partial response lasting nearly 3 years after receiving a selected neoantigen-reactive TIL product.
“For the common solid epithelial cancers — like colon and stomach and pancreas and uterus and ovary and prostate — the TILs will not work unless you actually select the cells that are reactive against the cancer,” Steven A. Rosenberg, MD, PhD, principal investigator of the study, told Medscape Medical News. “They turn out to be only about one in 1000 of those cells. When you select those cells, you begin to see responses.”
How Was the Study Designed?
This was a single-institution, nonrandomized phase 2 trial involving 91 patients with mismatch repair–proficient (microsatellite-stable) metastatic GI cancers, all of whom had progressed on standard therapies. Cancer types included colorectal, pancreatic, cholangiocarcinoma, gastric, esophageal, and ampullary cancers.
Participants first underwent surgical resection to harvest tumor tissue for TIL expansion. Patients were then assigned to three sequential treatment arms: Bulk, Unselected TILs (n = 18); selected neoantigen-reactive TILs (n = 39); and selected neoantigen-reactive TILs with pembrolizumab given immediately before infusion (n = 34). All patients also received lymphodepleting chemotherapy and high-dose interleukin (IL)-2.
The primary endpoint was objective response by Response Evaluation Criteria in Solid Tumors version 1.0. Safety and exploratory correlative analyses were secondary endpoints.
What Were the Key Findings?
Objective response rates for unselected TILs, selected TILs, and selected TILs plus pembrolizumab were 0%, 7.7%, and 23.5%, respectively.
Responses were seen across cancer types, with durable tumor regressions lasting up to 70 months. Adding pembrolizumab was critical to overcoming T cell exhaustion and improving efficacy.
“When we administered a checkpoint inhibitor like pembrolizumab — which in and of itself has no activity in these patients — response rates went up to 24%,” Rosenberg said.
All patients experienced at least one grade 3 or higher adverse event, but toxicity was largely attributed to the preparative chemotherapy and high-dose IL-2 rather than the TILs themselves. Serious events occurred in 30% of patients, with 10% requiring critical care. One patient died of adenoviral hepatitis. Pembrolizumab-related toxicities included steroid-responsive pneumonitis and treatment-refractory colitis. Most lower-grade events — such as nausea, diarrhea, fatigue, and electrolyte abnormalities — were consistent with known effects of the conditioning regimen.
“Ten percent of patients in the clinical trial required critical care support,” Lindsay M. Hannan, MD, MSc, a GI medical oncologist at the Winship Cancer Institute of Emory University, Atlanta, told Medscape Medical News. “I do not have specific information regarding the critical care needs of the patients, but this is concerning for excessive toxicity and tolerability in an already compromised population.”
How Far Is This Therapy From FDA Approval?
“These are very preliminary experimental results,” Rosenberg said. “It’s certainly going to take well over a year, or several years, before the FDA is ready to look at enough patients.”
Based on the study population, the most likely future indication would be for patients with treatment-refractory, mismatch repair–proficient GI cancers, including microsatellite-stable colorectal cancer.
To improve outcomes, Rosenberg’s team is exploring ways to streamline the TIL manufacturing process, better identify which T cells are most likely to drive clinical responses, and target more neoantigens per patient.
“The more different antigens that we target, the more likely we are to see a response,” Rosenberg said. “When we targeted three or more different antigens, our response rates were over 40%.”
Although the expansion phase of the trial is ongoing, recent federal funding restrictions and administrative changes at the National Institutes of Health are hindering progress.
“It’s certainly slowing us down,” Rosenberg said, describing challenges with hiring new employees, losing others to layoffs, and procuring necessary materials.
How Could These Findings Affect Clinical Practice?
“This represents a real shift in how we approach immunotherapy in GI cancers,” said Mark O’Hara, MD, associate professor of medicine, hematology-oncology, at the Hospital of the University of Pennsylvania, Philadelphia, in an interview.
“These are early results from a single institution, and further validation is needed,” he noted. “But if these data hold up, it could offer a treatment option where none currently exist.”
“The study highlights the potential of TILs in the treatment of gastrointestinal cancers,” said Hannan. “The presence of neoantigen-specific T cells in peripheral blood suggests that selective TIL-based therapies could be a viable strategy to treat some patients with treatment refractory disease.”
What Are the Potential Advantages and Disadvantages vs Current Treatment Options for GI Cancers?
“This approach allows for the identification and expansion of T cells that recognize patient-specific neoantigens, potentially enabling a more targeted and effective therapy,” O’Hara said. “The fact that some of these patients had prolonged benefit is particularly encouraging.”
On the other hand, both outside experts raised concerns about practical implementation.
“From a logistics perspective, the process is time-consuming, labor-intensive, and likely expensive, especially compared to treatments like chemotherapy or monoclonal antibodies,” Hannan said. “Not all patients are candidates, especially those without accessible tumor tissue, those without appropriate neoantigens or those who are too ill to wait for the cell expansion process.”
This study was supported by the Intramural Research Program of the National Institutes of Health, National Cancer Institute, Center for Cancer Research, with additional support through a collaborative research and development agreement with Iovance Biotherapeutics. The authors and O’Hara, and Hannan reported no competing interests.