Immunotherapy Combos Continue to Excel in NSCLC, But Patient Selection Must Be Optimized

Immunotherapy combinations have shifted from investigational regimens to novel frontline treatment options in non–small cell lung cancer (NSCLC), according to Michael Shafique, MD, who added that experts in the field are hopeful that ongoing research efforts will continue to optimize these therapies to ensure they’re getting to the patients most likely to respond to them.

“A few years ago, there were a slew of immunotherapy trials, but I think we’re starting to see now that immunotherapy has now been established as a frontline treatment option for most patients,” Shafique said. “But we’re seeing some exciting developments in the targeted space. In the future, the 2 biggest things we’re going to be seeing are advances in the targeted space, and more of an optimization of immunotherapy approaches.”

In an interview with OncLive® during the 2020 Institutional Perspectives in Cancer webinar on lung cancer, Shafique, a medical oncologist in the Department of Thoracic Oncology at Moffitt Cancer Center, highlighted examples of this groundbreaking research of immunotherapy combinations in NSCLC.

OncLive®: Which recent data have you found to be the most impactful in NSCLC?

Shafique: There are 3 major [recent] clinical trials that have impacted the standard of care for patients. Some immunotherapy trials included the evaluation of dual checkpoint inhibitor blockades, such as ipilimumab [Yervoy] combined with nivolumab [Opdivo], [or adding that doublet] to chemotherapy. [It is important to put in] perspective the standard of care that exists in result to some of the KEYNOTE trials, which combined chemotherapy with immunotherapy.

There is also a focus on targeted therapies. There are newer studies in the EGFR-mutated lung adenocarcinoma space and will discuss whether they would impact the current standard of care, which is osimertinib [Tagrisso] monotherapy.

Finally, there are a group of studies in small cell lung cancer [SCLC]. IMpower133 evaluated atezolizumab [Tecentriq] plus carboplatin and etoposide in patients with extensive-stage SCLC. Another trial is the CASPIAN trial, which evaluated carboplatin, etoposide, and durvalumab [Imfinzi]. This is another PD-L1 inhibitor, similar to atezolizumab. Then, in the second-line setting, lurbinectedin (Zepzelca) was approved off of a phase 2 trial in platinum-pretreated SCLC.

Looking at the combination of immunotherapy and chemotherapy, could you expand on the importance of longer follow-up from the KEYNOTE-189 trial?

Notably, [we need to be looking at] the long-term OS results in both KEYNOTE-189, along with some of the newer trials. What will truly help us differentiate between which therapy patients should receive in the frontline setting [can be decided based upon some of that long-term survival data].

As such, we’re going to be comparing a few trials with the final OS analysis in KEYNOTE-189, not just now, but also in the long run. There were not many new safety signals reported in the final analysis with pembrolizumab [Keytruda] plus chemotherapy. Thus, I believe this is a safe and effective regimen, and I think in a lot of cases, KEYNOTE-189 is sort of the benchmark for the chemoimmunotherapy combinations in nonsquamous NSCLC.

Despite the positive findings from the KEYNOTE-407 trial, what questions remain in squamous NSCLC?

[We need to be looking at what] the OS data are and the expected response rates [with this combination]. We will then contrast that with some of the newer data and at least raise the question of whether these benchmark trials should be used for most of our patients with no or low PD-L1 expression, or are they for other groups of patients in whom we should be thinking about using nivolumab and ipilimumab.

The community folks are comfortable utilizing [the KEYNOTE-407] regimen and they know what to expect in terms of response and adverse effects, but I think some the major questions that arose from the 2020 ASCO Virtual Scientific Program, from folks in the community, are: “What do we do with the dual checkpoint inhibitor trials, and how should we utilize CheckMate-9LA data?” How does this affect the KEYNOTE studies?

CheckMate-227 trial led to the FDA approval of dual immunotherapy in frontline NSCLC. How are you using these data in practice?

Dense data came out of the CheckMate-227 and CheckMate-9LA trials. The most important thing from CheckMate-227 is that it evaluated ipilimumab and nivolumab versus chemotherapy, and there was also an arm that contained chemotherapy plus nivolumab.

Regarding the OS benefit in the dual immunotherapy combination group compared with chemotherapy, the median OS was assessed in patients with at least 1% of PD-L1 expression. The median OS was about 17 months, which was statistically significant in an all-comer population. Unfortunately, this trial was not powered to seek OS in the patients who did not express PD-L1. Even so, there was an almost numerically identical median OS in that group, which was a little more than 17 months as well.

Ultimately, we learned that this regimen appears to be effective for both patients with PD-L1 expression and, those who were PD-L1 negative; it’s a chemotherapy-free regimen.

Some patients and providers find this appealing, especially if they’re worried about the toxicities associated with chemotherapy. There is still toxicity associated with the dual immunotherapy combination, so there is still a significant discontinuation rate. Of the patients who responded, their responses are very, very durable. These patients remain in partial response or complete response.

How do these data compare with those of the CheckMate-9LA trial?

Whenever you're administering immunotherapy alone or with chemotherapy, there's a small fraction of patients who seem to do better when they receive chemotherapy upfront. Although, in the long run, patients do better with immunotherapy. During this trial, we wanted to determine how to salvage those patients who are otherwise not responding quick enough up front to immunotherapy. Additionally, we wanted to minimize the amount of chemotherapy that we’re giving patients with up-front treatment.

As such, the idea of CheckMate-9LA was to give 2 cycles of chemotherapy with the first 2 cycles of immunotherapy. Similar to KEYNOTE-189, patients received a chemotherapy plus immunotherapy combination; however, instead of 4 cycles of the combination, these patients received 2 cycles of histologically driven chemotherapy. Following this, patients received ipilimumab and nivolumab maintenance for the duration of the study.

Notably, we saw an effect of chemotherapy in this setting whereas we did not see the initial advantage of chemotherapy alone, and we do seem to preserve the overall effect of immunotherapy in patients. This a promising treatment strategy. Neither of these, CheckMate-227 or CheckMate-9LA, were directly compared with KEYNOTE-189 or KEYNOTE-407, so we don’t really know how they stack up against each other. I think that’s where a lot of the questions are going to come up—which patients are best for which regimen?

What other challenges are currently being discussed in NSCLC?

An issue that we are running into is that patients who have an initial response to immunotherapy but later on become refractory or resistant to it. We don’t have a great understanding of the resistance mechanisms to immunotherapy treatments. We need to develop a greater understanding of why this occurs. By doing this, we may be able to expand the pool of patients who have durable responses to immunotherapy.

This is part of the evolving understanding of molecular biology and it allows us to develop more targeted options—not just TKI oral therapy, but antibody-drug conjugates that are targeting certain receptors that allow for more targeted delivery of chemotherapy to patients. A deeper understanding of molecular biologies is really going to play a role in the future.