Welcome to CME on ReachMD. This activity, entitled “Keeping Pace in Women’s Cancers: Updates in Triple-Negative Breast Cancer,” is provided by AGILE and is supported by an independent educational grant from Merck.
Before starting this activity, please be sure to review the disclosure statements as well as the Learning Objectives.
Here is Dr. Javier Cortés.
Metastatic triple-negative breast cancer, or TNBC, is far more aggressive and deadlier than other cancers and has few effective treatment options. As a result, metastatic TNBC presents a complex clinical challenge to physicians. How will emerging immunotherapies shift the treatment paradigm for TNBC?
This is CME on ReachMD, and I am Dr. Javier Cortés.
I’m Dr. Peter Schmid, and I look forward to discussing those important topics with you this afternoon.
So, Peter, to start us off, can you give us a brief overview of the biology of TNBC? How, in your opinion, do advances in the analysis of biology and heterogeneity assist in determining treatments?
Yeah, thank you, Javier. Our understanding of triple-negative breast cancer has increased dramatically over the last few years. But within triple-negative breast cancers, there are several subtypes. Depending on the classification, we choose possibly 4, possibly 6, and this is based on gene expression profiling, distinct subgroups with very different biology. We are also going forward by – maybe not use these classifications. But I think they’re helpful for us to understand how heterogenous triple-negative breast cancer is, but also to help us identify possible therapeutic strategies. There’s a whole range of new therapeutic targets that have been identified over the last few years and will be identified going forward. And I’m pretty optimistic that using those targets, we will also start redefining different subgroups of triple-negative breast cancer and tailor the treatment strategies according to those subgroups.
One of the examples is PD-L1 expression, which has been identified as a marker for immune therapy sensitivity, where we now have a new standard of care in the use of immune therapy in the first-line setting of triple-negative breast cancer. But there are other examples that are being developed as we speak, where we develop new targeted therapies for distinct subtypes of triple-negative breast cancer.
Now, Javier, when we first started developing immune therapy, we started with single-agent immune therapy, about 6, 7, 8 years ago. Would you mind just running us through the data with those single-agent immune therapy trials?
Well, thanks, Peter. Yes, I do know very well different immune checkpoint inhibitors have shown to have some activity in triple-negative breast cancer when used as monotherapy. In brief, I think, at least in my opinion, both pembrolizumab and atezolizumab have shown the most interesting activity in this group of patients. Overall response rate has been in the range of 4% to 7% in second-line or beyond, and in the range of 25% in the first-line setting. What is, in my opinion, more interesting, is the duration of response. Only a group of patients responded to these agents, but these patients could have very long-term outcomes, which in my opinion, is very, very interesting.
The most important trial, nevertheless, is the KEYNOTE-119 in this setting. As you know, these are second-, third-line clinical trial, with pembrolizumab compared with physician’s choice. The primary endpoint was overall survival. Unfortunately, overall survival was not superior for those patients who received immunotherapy compared with chemotherapy, but there was a very interesting exploratory analysis for those patients with high expression of PD-L1 measured with a CPS [combined positive score] of 20 or more. In this group of patients, again, there was a subgroup analysis, exploratory analysis, that survival was clearly more interesting in favor of pembrolizumab. So in my opinion, immunotherapy is there – might fail a group of patients, but very limited. And this is why, in my opinion, combining immunotherapy with chemotherapy is much more interesting.
For those just tuning in, you are listening to CME on ReachMD. I am Dr. Javier Cortés, and here with me today is Dr. Peter Schmid. We are discussing the role of immunotherapies in the treatment of metastatic triple-negative breast cancer.
So we have been talking about mono-immunotherapies. Now, Peter, building on that, how do we determine if a patient would be eligible for combination immunotherapies?
So the first two registrational studies, KEYNOTE-355 and IMpassion130, allowed patients regardless of PD-L1 expression in the trial. These are two first-line trials, first-line metastatic triple-negative breast cancer. In IMpassion130, patients received nab-paclitaxel as a chemotherapy backbone and then were randomized between atezolizumab or placebo in combination with chemotherapy. In KEYNOTE-355, investigators could choose between 3 chemotherapy options – conventional paclitaxel, nab-paclitaxel, or gemcitabine and carboplatin – and those patients were then randomized to be pembrolizumab in combination with chemotherapy or placebo.
Both trials show remarkably similar results. Both trials were positive for progression-free survival. In the ITTs, or intention-to-treat population, in all comers, showing the clear benefit from the addition of the immune checkpoint inhibitor. However, when we looked at those data in more detail, in pre-planned analysis, it became clear that the patients who derive a benefit are patients with PD-L1-positive tumors, whereas patients with PD-L1-negative tumors don’t seem to derive benefit from the addition of immune checkpoint inhibitors. PD-L1 expression was assessed differently in those two phase 3 trials. In KEYNOTE-355, the CPS score was used with a cutoff of 10, and this is based on an antibody called 22C3. In IMpassion130, an antibody called SP142 was used with a cutoff of 1%.
Now what is interesting is both assays describe a relatively similar population of around 40% of all triple-negative breast cancer patients. They overlap, largely by about 75%, although not completely. But it is clear now, following those two phase 3 trials, that patients in a first-line setting with metastatic triple-negative breast cancer benefit from the addition of immune checkpoint inhibitors, pembrolizumab or atezolizumab, but only if they’re PD-L1-positive tumors. In IMpassion130, we could also show with long follow-up that patients have a substantial survival benefit with an overall survival in the PD-L1-positive subgroups of more than 2 years. We’re still waiting for the survival data of KEYNOTE-355, but at the moment, all data seem to be very similar to what we saw with IMpassion130.
So, Peter, you have described very nicely how pembrolizumab/atezolizumab in combination with chemotherapy improves long-term outcomes, basically for PD-L1-positive patients. But it was surprising, to me at least, the results we saw from the IMpassion131 – so paclitaxel plus atezolizumab – did not behave in a similar way as nab-paclitaxel and atezolizumab or chemotherapy plus pembrolizumab. Do you have any feeling about why this trial failed to demonstrate an improvement in progression-free survival?
Yeah, Javier, that’s a really difficult question, and we can only speculate. IMpassion131 was, in terms of the design, very similar to IMpassion130, although the trial is smaller than IMpassion130, with around 650 patients. When we first designed IMpassion130, we deliberately chose nab-paclitaxel as a chemotherapy because back in 2013-2014 when we designed the trial, we were a little bit anxious about the use of steroids, which is generally required if you use conventional taxanes. But in IMpassion131, we wanted to test whether we could see similar results with paclitaxel – conventional paclitaxel – compared to nab-paclitaxel in IMpassion130. And the short answer is we failed to show this. There was no benefit in progression-free survival or overall survival from the addition of atezolizumab to conventional paclitaxel.
The question now is why is that? And the jury is still out. We haven’t got a definitive answer. There is certainly some small changes in the trial design that may have contributed to this. The trial is smaller, about 200 patients smaller. It used the 2:2:1 randomization, so the control arm in IMpassion131 is relatively small. We also see an unusually good survival in the control group in IMpassion131 – control group treated with just the paclitaxel and placebo for more than 2 years, which is unheard of in this group of patients and may be a result of the heterogeneity of triple-negative breast cancer. But also, we are aware that patients in IMpassion131 received steroids – comedication – and we don’t know whether that had an impact or not. So what we don’t know, at this point in time, was there something not working out in that trial? Was, really, the combination of paclitaxel and atezolizumab suboptimum? For now, I would put a slight caveat behind using that combination.
On the other hand, in KEYNOTE-355, Javier, as you know and as you presented, patients could use paclitaxel or nab-paclitaxel or gemcitabine and carboplatin. And the most recent data that were presented in San Antonio, we saw a breakdown of the benefit for patients who had paclitaxel and nab-paclitaxel, and there didn’t seem to be a difference between those 2 subgroups. So it is unclear whether this was just a glitch in a trial with IMpassion131, but as those data are there, and are in contrast to IMpassion130, and in contrast to KEYNOTE-355, I personally will be cautious about using conventional paclitaxel in this setting rather than nab-paclitaxel.
Yeah, I think you are totally right.
Are there any barriers at the moment for using those checkpoint inhibitors in breast cancer, but also for treatment of metastatic triple-negative breast cancer in general? And what can we as clinicians do to overcome these barriers?
Well, I think of course we have barriers, as always in life. But I would mention maybe 2 or 3 of them. The first one is that we have to avoid – we have to try not to have patients with metastatic triple-negative breast cancer, and in my opinion, the best way to do that is to optimize the treatment for patients with early breast cancer. So, you know, Peter, you have been there – lead investigator of the KEYNOTE-522 with pembrolizumab in combination with chemotherapy in the early breast cancer setting, in the neoadjuvant setting. So I think that if we are able to decrease the probability for our patients to have metastasis in the future, optimizing treatments in the early breast cancer setting is, in my opinion, the best we can do for them. So let’s see if in the future with the long-term outcome data, we also have the opportunity to use these agents in this setting.
I think that also some barriers, of course, are how to optimize who are going to be the patients who will benefit more. We are talking about biomarkers. We have today the PD-L1 data with different platforms from different drugs – atezolizumab and pembrolizumab – but we also have other biomarkers. We are talking about microbiota. We are talking about liver metastasis, LD8, I don’t know – we have plenty of possibilities to try to optimize the group of patients who will benefit more.
I mean, one of the interesting points you brought into this discussion is the introduction of immune therapy into early triple-negative breast cancer therapy. And what we have seen in two randomized phase 3 trials is a substantial, and in my opinion, clinically meaningful increase in the pathological complete response rates by between 13% and 17% in those two studies. What’s also interesting there is that that benefit seems to be even more pronounced in patients with higher-risk disease. In fact, if you look at patients with node-positive disease versus patients with node-negative disease, if we treat them with chemotherapy alone, there’s a massive difference in path CR rates. But if we treat those patients with chemotherapy and immune therapy in the neoadjuvant setting, there’s no difference in path CR rates anymore between node-positive patients and node-negative patients. But really most importantly, and going back to what you said about barriers and biomarkers, what is really interesting, the PD-L1 expression doesn’t seem to be a predictive biomarker of response in the neoadjuvant setting, whereas as we discussed earlier, it is a very highly predictive factor in the metastatic setting.
So, Peter, you know, it is always absolutely great, fascinating conversation to meet with you, to talk to you, and to share clinical trials and experience with you. But we have to conclude, so can you please share with the audience just a 20-second take-home message?
We are starting to make a real difference in triple-negative breast cancer. Immune therapy has been a massive first step forward, improving overall survival in metastatic triple-negative breast cancer and also path CR rates in early triple-negative breast cancer. Our understanding of the biology of triple-negative breast cancer is allowing us to develop more targeted therapies.
I only can echo all your words, Peter. Unfortunately, that’s all the time we have today, so I would like to thank you all for listening in, and thanks a million, Peter, for joining me and for sharing all your great insights. It was great speaking with you today, as always. Thank you, my friend. Thank you, everybody. Stay well and goodbye.
Thanks also from my end. It’s always a pleasure, Javier, connecting with you. Thank you everyone, for listening in. Look forward to seeing you face-to-face in the near future. Stay safe.
You have been listening to CME on ReachMD. This activity is provided by AGILE and is supported by an independent educational grant from MERCK.
To receive your free CME credit, or to download this activity, go to ReachMD.com/AGILE. Thank you for listening.