Frontline I-O Treatment for mCRC

Tanios S. Bekaii-Saab, MD: We know that MSI-high is a good predictor for pembrolizumab and other IOs [intraosseous infusion] and we know that many patients could be technically cured, but KEYNOTE-177 was published in the New England Journal Medicine and that moved it to the first-line therapy. Tell us about it and what did that do to your clinic?

Cathy Eng, MD, FACP, FASCO: KEYNOTE-177 was the big phase 3 trial specifically in treatment naïve patients that were MSI-high and compared with standard chemotherapy imposed patients that received chemotherapy were allowed to crossover. The primary end point was progression-free survival [PFS]. They demonstrated that as they presented that last year it was double that, it was over 16 months versus the control arm of little over 8 months.

What does it do for our clinic? Many of us would probably choose IR [infrared] therapy as a frontline treatment for our newly diagnosed patients, but there are some intriguing data. When you look at the KM [Kaplan-Meier] curves for the PFS, about a third of patients that received IR therapy did not fare as well as the patients that received standard chemotherapy. Why that is? We don’t have all the answers yet. It's quite intriguing. Overall these patients did well in those patients that did respond. These were durable, long-lasting responses. The quality of life has been presented at ESMO as well and that was superior to chemotherapy.

Many people are going to move to single-agent pembrolizumab for newly diagnosed patients. The response rate was 43% versus 33% for standard chemotherapy. I am intrigued by CheckMate 142, which was a single-arm study for treatment IO patients looking at nivo [nivolumab] and ipi [ipilimumab]. The response rate was extremely high at 69%. That was updated this year. I am intrigued by that, but once again, single-arm phase 2 versus the large phase 3.

Tanios S. Bekaii-Saab, MD: How long is too long, and how short is too short for pembrolizumab in this setting?

Cathy Eng, MD, FACP, FASCO: All of the trials, when they were originally designed, allowed for a maximum of 2 years of treatment on the IO therapies. Every single IO trial allowed a maximum of up to two years and then you could ask if you could continue. This came up at a case session I had with some fellows at the University of Toronto asking, at what point do stop? Some of these patients have such beautiful responses. Do you still want to take them to surgery when they've had complete fibrosis of their disease and have had amazing responses? These are the questions we don't have answered at this time because IO therapy, at least in gastrointestinal a, especially in colorectal, is so new. I don't have a great answer for you in regard to a response because right now it's very subjective once you get past that 2-year interval if you've had a beautiful response.

Tanios S. Bekaii-Saab, MD: Are you stopping at 2 years?

Cathy Eng, MD, FACP, FASCO: I have not had the luxury of a patient being on it past 2 years thus far, except in a patient in 1 of my anal carcinoma trials. My understanding is that patient, though, continued it intermittently. Actually, at John Strickler, MD,'s institution [Duke Cancer Institute].

Joleen M. Hubbard, MD: I've had 2 patients that have had a complete response to single-agent pembrolizumab. I didn't know when to stop, I stopped after 2 years. Both of them had their complete response within about 9 months of therapy. I didn't know what to do at that point, I kept going with the treatment and then discontinued for sure after two years. I could have stopped earlier, I was a little nervous and I went by what the trials did. Since then, neither one has recurred but if they did recur I wouldn't hesitate to put them back on immunotherapy.

Tanios S. Bekaii-Saab, MD: If you look at the curves from KEYNOTE-177, a major inflection point is approximately 1 year. You asked the question whether beyond 1 year makes sense. I have patients who had, for whatever reason, to drop approximately 9 months and 11 months and remained NCR [near complete response] 5 1/2 years later. They were on the original study. The intriguing part is that you get your response, but your maximum response is not showing before another 6-plus months, at least radiographic response. Sometimes you take those patients to the OR [operating room] because they have a stubborn liver lesion that is not going away and its pathologic complete response. We don't know how many actually end up in pathological complete response that still have these tumors radiographically present.

Transcript Edited for Clarity