Milind Javle, MD: I’m going to jump right into 1 of the main topics of discussion, which is the role of immune checkpoint inhibitors. I’m going to give you a brief overview of the TOPAZ-1 study for the group, but I’m going to rely on the experts to dive in deep to help us understand the data better. TOPAZ-1 was the first study after a decade, a phase 3 trial. We were hungry for a positive study after the ... modest survival benefit that we get as Mark outlined. TOPAZ-1 was a large multicenter study of 685 patients, about 340 of whom got gemcitabine-cisplatin plus durvalumab. The other about 340 got gemcitabine-cisplatin. They got gemcitabine-cisplatin and the study therapy, the durvalumab or placebo, for 6 months or 8 cycles. After that they got durvalumab as a maintenance agent or placebo.

The study was positive from all fronts. The overall survival was a primary end point. The hazard ratio for overall survival was 0.80 and for progression-free survival [PFS] was 0.75. The response rate was higher in the study arm, about 26%, with gemcitabine-cisplatin-durvalumab vs 18% with gemcitabine-cisplatin. Very interestingly, at the end of 2 years, 10% were alive in the study arm with gemcitabine-cisplatin and placebo vs 25% with the gemcitabine-cisplatin and durvalumab. There seems to be a tail in the curve, where patients who receive checkpoint inhibitors did better. We’re excited about these results. It’s being incorporated and practiced based on NCCN [National Comprehensive Cancer Network] Guidelines. We’re hoping for an FDA approval very soon.

I’m going to first ask a higher-level question. Mark, why should we consider immune checkpoint inhibitors in biliary tract cancer when you said this was a cold tumor to begin with? Why is there rationale to investigate checkpoint inhibitors at all?

Mark Yarchoan, MD: The simple answer is that immune checkpoint inhibitors have been thrown at every tumor and have transformed the management of virtually every tumor. After some easy wins in tumors like melanoma and lung cancer, we’ve gone after the hardest tumors. Even in immune-resistant tumors like cholangiocarcinoma, there clearly are responses to immunotherapy. Unfortunately, we don’t know who those responders are. We don’t have effective biomarkers in cholangiocarcinoma, so it makes sense to offer this to all patients.

The other thing that’s worth mentioning is the potential for synergy between chemotherapy and immunotherapy. Does chemotherapy induce immunologic cell death? Does immunotherapy deepen those responses? There were many reasons to be hopeful that this trial would be positive. Sure enough, as you mentioned, this was clearly a positive study for overall survival PFS and response rate, even if the overall benefit was more modest than we’ve seen in other tumor types.

Milind Javle, MD: Thank you, Mark. Anjana, from the perspective of underlying liver disease, as a hepatologist, do you see a rationale for using immunotherapy in these tumors?

Anjana Pillai, MD: I rely heavily on my oncologist for the interpretation of many of these studies, but I don’t see a problem with using them if you use it again if you’re looking at patients with compensated disease. This is where the multimodal multidisciplinary part of treating these cancers come in. If someone has preserved liver function, ideally Child-Pugh A or even early B, doesn’t have autoimmune hepatitis undergoing a flare or a refractory or that’s on treatment or high levels of treatment, then that’s reasonable to try. With everything else, we talk to these patients about the risk and benefit with caution, but if we never try, the outcome is much less. Every additional month matters in someone’s life if you preserve quality and as long as you’re approaching it with that in mind. That’s a long-winded answer, but I don’t see any reason not to try in those with preserved function and no obvious contraindications.

Milind Javle, MD: Ruth, do you have a privilege situation because you treated a lot of patients on this trial? You’re 1 of the lead investigators, so help us understand the disease a little better. There are 3 types of biliary tract cancers, as Flavio mentioned. What’s the relative benefit? There was a recent poster for ESMO [European Society for Medical Oncology] in terms of regional variation: Asia vs the rest of the world. Help us understand the subgroups and ethnicities and how we interpret these data given the diversity of these cancers.

Aiwu Ruth He, MD: Why did we do a subgroup analysis? First, this study design has 2 stratifications. One is by disease status. It’s initially diagnosed advanced-stage or recurrent disease. The other is anatomic location. Those are predefined in the study at the beginning. The reason is because biliary cancer is a very heterogeneous disease, as the team has mentioned, and each anatomic subtype can have different risk factors, etiology, presentation, or prognosis. That’s why the data were in the primary analysis. The hazard ratio for overall survival favored the combination across the locations or primary locations consistent with a full analysis set.

There are 2 things that people are very interested in looking at. Even though this is a global phase 3 study, if you try to look at 3 anatomic locations, in different regions, of course there’s a certain subset. You won’t have enough events to calculate the hazard ratio. The team is trying to calculate the hazard ratio for all types of combinations. The first question to address is about the Western population. The 1 criticism is the initial study, a phase 2 study from Korea, which has shown very promising results. Because this is expected, we would ask if is this a Western-Eastern difference. The hazard ratio presented at the TOPAZ-1 is 0.80, and the initial phase 2 data showed a higher overall benefit. The hazard ratio for anatomic location each time were calculated—you combine European and North American data. That will represent a Western population. Then we see a hazard ratio show the benefit or support, show the benefit of durvalumab ... across the primary tumor location.

The other question asked about is gallbladder cancer. In the initial data presentation, gallbladder cancer has a higher hazard ratio. As you know, gallbladder cancer has a unique geographic distribution based on epidemiology. The hazard ratio was reported higher in a subgroup analysis from the TOPAZ-1 study. That’s another reason why this detailed subgroup analysis was evaluated. Therefore, when you look at gallbladder cancer, if you look at all geographic area, it favors combination therapy. Though in South America, the hazard ratio was not calculated because the events were very small. If you look at the number in South America, patients with gallbladder cancer in the control arm outperform the historic control. That’s the reason it initially brought up the hazard ratio of the whole group. That’s the problem when you have to divide this number into many subgroups. When we look at the benefit, it looks like it favored all the primary tumor locations in all geographic areas. That’s overall survival. But if you look at PFS, this goes along with the overall survival.

Transcript edited for clarity.