Results from a phase 2 trial of temozolomide or temozolomide plus capecitabine showed promising efficacy of the regimens for patients with advanced pancreatic neuroendocrine tumors.
Temozolomide (Temodar) as monotherapy or with capecitabine was analyzed in a phase 2 trial (NCT01824875) of patients with pancreatic advanced neuroendocrine tumors (pNETs), with results from the updated analysis presented at the 2022 American Society of Clinical Oncology (2022 ASCO) Annual Meeting by Pamela L. Kunz, MD.1
The trial was designed to determine the safety and efficacy advantages of temozolomide in this population. Based on temozolomide’s ability to induce DNA methylation leading to DNA damage and cell death and its potential synergistic activity with capecitabine, the combination was investigated as a potential option for pNETs. Previously reported data showed that the combination resulted in superior progression-free survival (PFS) vs temozolomide alone.2
“Capecitabine and temozolomide is a standard treatment for patients with advanced pancreatic NETs,” said Kunz. “It’s been adopted into several guidelines, but [these results presented at 2022 ASCO] will strengthen that,” Kunz, associate professor of Internal Medicine; director of the Center for Gastrointestinal Cancers at Smilow Cancer Center and Yale Cancer Center; chief of Gastrointestinal Medical Oncology; and vice chief of Diversity, Equity, and Inclusion at Yale School of Medicine, said in an interview with CancerNetwork®.
Kunz also spoke about the rationale for creating this trial and reviewed efficacy data. As with its activity in glioblastoma, MGMT could represent a predictive marker in this cancer, which Kunz addressed in her review of the data. Additionally, she discussed the safety profile reported in the updated analysis and where research efforts should be focused in the future.
Kunz: The phase 2 trial was a National Clinical Trial Network trial through ECOG-ACRIN that looked at temozolomide and temozolomide plus capecitabine for patients with advanced pancreatic NETs. The reason we thought about doing this study is that there are several patients with pancreatic NETs that need objective responses and many of our currently FDA approved treatments yield stability and can control the cancer but don’t shrink it. Temozolomide is an oral alkylating agent, and a prior alkylating agent called streptozotocin, which is given intravenously, has been FDA approved since the 1980s. However, it’s associated with a high [adverse] effect rate, and temozolomide is oral, it has fewer [adverse] effects, and it could cross the blood-brain barrier. It has several advantages over streptozotocin and there were some preliminary data based on a small retrospective and prospective studies that suggested that it resulted in high response rates. There’s also a suggestion that methyl guanine, methyl transferase, which is a DNA repair enzyme, may be associated with a response. We also wanted to study that and see if it was associated with a response to temozolomide.
There were 144 patients randomized to receive either temozolomide or temozolomide and capecitabine. We presented our initial interim analysis at ASCO 2018 [analyzing] progression-free survival [PFS], which was the primary end point. We did meet our primary end point with a median PFS of 22.7 months in the combination arm and 14.4 months in the single-agent temozolomide arm with a hazard ratio of 0.58 [95% CI, 0.36-0.93; P = .022]. That was the final PFS because we met it at the interim analysis. The updated analysis that we presented at this meeting was of overall survival [OS]. We did not reach a statistically significant difference in OS between the arms. This was likely due to a long OS in this patient population of low-grade NETs. The median OS was 53.8 months in the temozolomide arm and 58.7 months in the combination arm [HR, 0.82; 95% CI, 0.51-1.33; P = .42]. There was still a clinically meaningful 5-month difference, but it did not meet statistical significance. This is a real challenge and in the NET clinical trial space, we don’t have agents that typically meet the OS end point. That’s why PFS is a preferred end point for those studies. The temozolomide arm was associated with a 34% response rate and the combination arm [yielded] a 40% response rate. Both were robust response rates. The study was not designed to detect a difference between those 2 arms as that was not the primary end point.
The other exciting data we presented were around MGMT. We looked at MGMT by both immunohistochemistry and by promoter methylation, so 2 different methodologies, with the hypothesis that MGMT deficiency predicts for or is associated with response. We did in fact find that. The study itself was not designed to look at MGMT as a predictive biomarker because temozolomide is used in both arms. We were able to show that it’s associated with response with both immunohistochemistry and by promoter methylation methodologies. The other thing we looked at was the relationship between promoter methylation and immunohistochemistry. We found that all the patients who were promoter methylation positive were also immunohistochemistry positive, but the reverse was not true. This suggested that promoter methylation was not sufficient as a biomarker for pancreatic NETs to predict or be associated with MGMT deficiency. Future studies are needed.
The combination arm was associated with a higher rate of [adverse] effects by almost double. When we looked at all the treatment-related adverse effects in greater than or equal to 5% of patients, they were mostly hematologic or gastrointestinal, like nausea and vomiting. Capecitabine and temozolomide was associated with a higher grade 3/4 toxicity rate, with almost double the adverse effects compared with temozolomide alone. When I’m looking at patients and making a treatment decision, patients who have comorbidities or who are elderly [could receive] temozolomide alone as an option.
Temozolomide can safely be a comparator arm in future studies for pancreatic NETs. I have served as the chair of the Neuroendocrine Tumor Task Force for the NCI [National Cancer Institute]. We’ve been thinking about this question and have 1 study ongoing and 1 study in development that both use capecitabine and temozolomide. The first is a SWOG study that looks at patients with high-risk, pancreatic NETs who are able to [undergo] resection. These are localized, and they are randomized to receive 4 months of capecitabine and temozolomide post operatively vs placebo. That’s exciting because the current standard of care is to give no adjuvant treatment for that patient population. The other is a study in advanced pancreatic NETs and it’s not yet open. This is a study through the Alliance [for Clinical Trials in Oncology], and this is a randomized phase 2 study looking at lutetium177 dotatate [Lutathera] vs capecitabine and temozolomide.