Adding Celecoxib to FOLFOX Does Not Demonstrate DFS Benefit in Colon Cancer

Jeffrey A. Meyerhardt, MD, MPH, FASCO

The addition of celecoxib to standard chemotherapy did not significantly improve disease-free survival (DFS) or overall survival (OS) in patients with colon cancer, according to Jeffrey A. Meyerhardt, MD, MPH, FASCO, who added that the next steps for research include determining whether certain subpopulations might benefit more from this approach over others.

In the randomized, phase 3 CALGB/SWOG 80702 trial, investigators sought to assess the efficacy of adding celecoxib to 5-fluorouracil, leucovorin, and oxaliplatin (FOLFOX) in patients with stage 3 colon cancer who had at least 1 pathologically confirmed positive lymph node or no regional lymph node disease.

Results showed a 3-year DFS rate of 76.3% with celecoxib versus 73.4% with placebo (HR, 0.90; 95% CI, 0.77-1.04; stratified log rank P = .1587). The 5-year OS rate with celecoxib versus placebo was 84.0% versus 81.8%, respectively (HR, 0.89; 95% CI, 0.73-1.07; stratified log rank P = .2204). No significant interactions were detected by age, stage, risk group, concurrent low dose aspirin use, sex, race, ethnicity, baseline ECOG performance status, assigned duration of adjuvant FOLFOX, body mass index, or tumor location.

“We’re currently conducting an analysis where we are studying molecular changes in tumors to see whether certain subpopulations exist that may benefit [from this approach] more than others,” said Meyerhardt, a professor of medicine at Harvard Medical School. “Again, there could be no effect, or it could be that this approach is only effective in a certain subpopulation. We can now look into this further, given the size of the study.

In an interview with OncLive, Meyerhardt, who is also clinical director of the Gastrointestinal Cancer Center; deputy clinical research officer; Douglas Gray Woodruff Chair in Colorectal Cancer Research; senior physician; and co-director of the Colon and Rectal Cancer Center, at Dana-Farber Cancer Institute further discussed the findings from the phase 3 CALGB/SWOG 80702 trial in patients with colon cancer and highlighted other ongoing research efforts.

OncLive: What was the rationale behind the phase 3 CALGB/SWOG 80702 study? 

Meyerhardt: A lot of data have been collected from observational studies, where investigators accrued information about subjects and followed them through any diseases they developed, as well as from randomized trials, which looked at both aspirin and COX-2 inhibitors. Patients who took those medications had a lower risk of developing colon polyps and data show that, in patients who already had colorectal cancer (CRC), the risk of subsequent polyps is reduced.

Some other observational studies were done where investigators followed patients with CRC who were taking either aspirin or a COX-2 inhibitor, usually for cardiovascular or arthritis reasons. Patients who consistently took them appeared to have a lower risk of disease recurrence and improved survival.

Ultimately, in medicine, we try to figure out whether something is truly helping through a randomized trial. The rationale of this study was to gather patients with stage 3 colon cancer, who already had surgery and positive lymph nodes, and to then give them chemotherapy. The second part of this trial was to look at the duration of chemotherapy, which was part of a larger international effort. We standardly give some patients 3 months, while others we give 6 months, depending on how severe their stage 3 disease is. We wanted to determine whether adding celecoxib for 3 years would further improve outcomes, reduce the risk of recurrence, and improve OS.

Did any previous trials show efficacy with the inhibitor to support its use in this patient population?

Trials were done for both celecoxib and a different COX-2 inhibitor rofecoxib, which was removed from the market because of cardiac issues in patients who had prior colon polyps, not necessarily colon cancer. These randomized trials showed a lower risk of subsequent polyps if patients received either celecoxib or rofecoxib. No recurrence data from a randomized trial were available for patients who had CRC. The phase 3 CALGB/SWOG 80702 was the first to examine these types of agents in patients with colon cancer.

Ongoing trials are examining aspirin, some of which are fully enrolled while others are still enrolling. The only study that looked at a COX-2 inhibitor in patients with colon cancer never completed accrual. Quite a few years ago, the VICTOR trial started accrual but, because of the cardiac issues found with rofecoxib, it closed early. The accrual goals that were needed to examine the question specifically in relation to potential cancer recurrence benefit were not met.

What were the goals of the phase 3 CALGB/SWOG 80702 study, and what patient population was examined?

A National Cancer Institute (NCI) sponsored study was conducted throughout the United States and Canada. The study included academic cancer centers and many community oncology practices, and it examined a fairly diverse population. We were specifically studying patients with stage 3 disease who previously underwent surgery and who had positive lymph nodes in their surgical specimen. These patients received standard chemotherapy. The specific goal of the study was to see whether the addition of celecoxib to chemotherapy would reduce the risk of recurrent disease and improve OS in these patients.

What findings were presented during the 2020 ASCO Virtual Scientific Program?

Unfortunately, the study did not show a statistically significant difference between the 2 treatments. A small difference in 3-year DFS was observed. Patients who received placebo had a 3-year DFS rate of 73.4%, while those who received celecoxib had a rate of 76.3%; this was a small difference of about 3%, but it was not statistically significant. We can't say that there was a difference between the 2, because it could still be by chance. Similarly, no difference in OS was observed either. We saw about an 82% to an 84% improvement in 5-year OS but, again, that was not determined to be statistically different.

Why do you believe there was no significant improvement in DFS or OS when celecoxib was added to chemotherapy?

The other important part of this trial is that we collected tumor specimens from the patients, such as blood samples. We also collected questionnaires regarding food and lifestyle. Some data show that these agents, COX-2 inhibitors as well as aspirin, may be more effective against certain types of colon cancers and that they have [efficacy in] certain molecular alterations.

What did the safety profile of celecoxib look like in this population?

We looked at the safety of receiving celecoxib while also receiving standard chemotherapy. Patients continued on celecoxib for up to 3 years. We also looked at celecoxib alone. In general, significant differences were not observed for most toxicities. A slight increase in high blood pressure was observed when patients received celecoxib during chemotherapy. Also, a slight increase in the number that relates to kidney function was also noted; however, a difference in kidney failure or anything like that was not observed.

Importantly, a significant difference in terms of cardiac events was not observed and that has been a concern with COX-2 inhibitors. Again, the reason rofecoxib was removed from the market was related to cardiac events. We didn't see a significant difference in cardiac events between the arms; both arms [had an] extremely low [number of cardiac events] and that difference was not determined to be statistically significant.

Were any unanswered questions left from this study?

Ongoing trials are looking at aspirin. It might be that aspirin has a different effect, so we’re awaiting those results. The other unanswered question is, “Could there be certain molecular features in colon cancer, where these agents may be more effective?” Those are studies that are ongoing within this trial.

Are you excited about any other ongoing trials?

Within this space, ongoing trials are using aspirin. Many questions still need to be answered to fully understand the biology of colon cancer, especially early stage colon cancer. We've collected many exposures so that we can answered these questions.

Another abstract is looking at nonoperative management in rectal cancer called the OPRA trial, which was led by a group at Memorial Sloan Kettering Cancer Center. The trial included approximately 10 or 11 institutions. Here, they helped define a patient population and a treatment regimen that could be considered for potentially avoiding a colostomy requiring surgery for rectal cancer. Those efforts are now being used to promote a larger trial within the NCI network that's pending evaluation to enroll.

Reference

Meyerhardt J, Shi Q, Fuchs CS, et al. Celecoxib in addition to standard adjuvant therapy with 5-fluorouracil, leucovorin, oxaliplatin (FOLFOX) in stage III colon cancer: results from CALGB/SWOG 80702. J Clin Oncol. 2020;38(suppl 15):4003. doi:10.1200/JCO.2020.38.15_suppl.4003