Eric S. Nadler, MD, discusses key data with immunotherapy in the treatment of patients with metastatic squamous non–small cell lung cancer.
Eric S. Nadler, MD
Chemoimmunotherapy has emerged as the frontline standard of care in patients with squamous non—small cell lung cancer (NSCLC), but work still needs to be done by way of determining effective biomarkers and identifying other effective combinations, said Eric S. Nadler, MD.
In October 2018, pembrolizumab (Keytruda) was FDA approved for use in combination with carboplatin and either paclitaxel or nab-paclitaxel (Abraxane) for the frontline treatment of patients with metastatic squamous NSCLC. The approval was based on data from the KEYNOTE-407 study, which demonstrated a 36% reduction in the risk of death with the addition of pembrolizumab to standard chemotherapy versus chemotherapy alone.1 The survival benefit was observed across PD-L1 subgroups.
In the phase III trial, treatment-naïve patients with metastatic squamous NSCLC were randomized to receive either carboplatin and paclitaxel or nab-paclitaxel plus pembrolizumab versus placebo followed by single-agent pembrolizumab or placebo. Objective response rate was 58% in those who received pembrolizumab versus 35% in the control arm. At a median follow-up of 7.7 months, median overall survival (OS) was 15.9 months versus 11.3 months in favor of the pembrolizumab arm. The addition of pembrolizumab also led to an improved median progression-free survival (PFS) of 6.4 months versus 4.8 months.
Beyond pembrolizumab, the addition of atezolizumab (Tecentriq) to chemotherapy also made headway in patients with squamous histology, although the agent has yet to receive approval for this indication. Updated data from the phase III IMpower131 study showed a median OS of 14.6 months versus 14.3 months in favor of atezolizumab plus carboplatin/nab-paclitaxel compared with carboplatin/nab-paclitaxel alone.2 Furthermore, the median PFS was 6.5 months versus 5.6 months in the chemoimmunotherapy and standard-therapy arms, respectively.
Although the data from IMpower131 are not quite as impressive as that seen with pembrolizumab, there is still time for the data to mature, said Nadler, a medical oncologist at Baylor University Medical Center and medical director of US Oncology Health Informatics and Internet Technology.
In an interview during the 2019 OncLive® State of the Science Summit™ on Non—Small Cell Lung Cancer, Nadler discussed key data with immunotherapy in the treatment of patients with metastatic squamous NSCLC.
OncLive: What are the biggest takeaways from the KEYNOTE-407 study?
Nadler: This was a very large, multinational study with over 150 participating centers. [For the study], you could use either carboplatin plus paclitaxel or carboplatin plus nab-paclitaxel. The majority of the patients, more than 50%, received nab-paclitaxel. These are data that echo what we knew before, which is that survival and response rates [are a little bit better] in the squamous cell population with that agent; however, nab-paclitaxel hasn't been used that much in the United States. That’s one thing that was interesting about the study.
The second thing is obviously the addition of immunotherapy. We had the nonsquamous data far before this, but pembrolizumab was FDA approved for use in the squamous population in October 2018. Really, what we saw was an improvement in PFS and OS that was demonstrable and measurable. The [positive] data were most clearly seen in patients who had high PD-L1 expression, and secondarily, in those who had intermediate PD-L1 expression. For the patients who had low PD-L1 expression, the survival benefit and hazard ratio were actually quite pronounced. With that being said, the PFS was a little less marked for these patients. We see that time and time again in these studies, whether it be in stage III disease or in nonsquamous NSCLC. We are seeing differential benefits, in terms of whether PD-L1 is greater 50%, between 1% and 49%, and less than 1%.
This has basically become the de facto standard-of-care regimen for frontline squamous NSLC. If it hasn't already become so in more than 50% of the United States marketplace, it will shortly. We are seeing a pretty substantial uptick [of its use]. Again, we tend to use carboplatin/paclitaxel in the United States, but both options are fine for this regimen.
The only other thing I'll mention is that the crossover rate was lower than one would anticipate. Only about 30% of patients crossed over to have immunotherapy, and when you actually looked at the intent-to-treat population, only about 40% crossed over to immunotherapy. Given that this trial was accruing at a time where we knew second-line immunotherapy was a very beneficial therapy, you would question why the rates were not even 50% in patients who moved on to second-line treatment. You would think the crossover rate would be at least 50% and probably higher than that. We don't have a great answer to this, but we definitely have to question why.
Is this regimen an option for patients with low PD-L1 expression or those who are nonexpressers?
Every patient is eligible to receive this regimen. However, the benefit we'd anticipate or expect may be markedly different if the patient has a PD-L1 status of 50% versus 90% versus 0%. With that being said, all patients should receive this regimen regardless of PD-L1 status. This is the FDA-approved combination—whether you use it with paclitaxel or nab-paclitaxel is up to the discretion of the physician. It wouldn't surprise me as data emerge, and we learn more about factors like tumor mutational burden (TMB), that there are some patients who have low PD-L1 expression who don't get this regimen upfront. Currently, all patients should receive this, but whether or not that will continue to be the case moving forward is unclear.
What are your thoughts on the IMpower131 findings?
The data with atezolizumab had a similar trial design [to the KEYNOTE studies] except that having paclitaxel or nab-paclitaxel was actually a trial randomization. The other thing is that there wasn't a crossover in this population. Patients were able to receive second-line immunotherapy, but crossover wasn’t allowed. The PFS benefit in this trial was significant in all regards, and [the benefit in terms of] OS was certainly there, just not to the same degree that we saw in the first study. The hazard ratios were all positive. Again, we see that the PD-L1 nonexpressers do not do quite as well, the response rates are lower. We are seeing this continued signal that PD-L1 less than 1%, or essentially nonexpressers, tend to derive less benefit from immunotherapy.
This trial did not quite have the same result in endpoint in either PFS or OS as the KEYNOTE studies did. This is not an FDA-approved regimen and we are not routinely using [this approach] in this patient population. That being said, as [the data] mature, and we see more data presented from this, we may feel very differently .
In general, how has the emergence of immunotherapy improved the outlook for patients with NSCLC?
Overall, by adding immunotherapy we are seeing a significant improvement in response rates compared with what we were used to seeing with chemotherapy alone. We're seeing OS that is markedly better than anything we've seen before; even when you add immunotherapy in the second-line setting, we are seeing survival benefit there. Our ability to deliver effective immunotherapy is much easier when you start it on day 1 rather than hoping to get it in the second-line setting. That is one of the interesting things that both of these trials demonstrate: the ability to get second-line immunotherapy is not always present.
We always think, "It's so easy to get, who wouldn't be able to get second-line therapy after progression on standard chemotherapy?" However, it turns out that these data clearly show that we are not always able to get second-line immunotherapy. That's part of the reason why these survival benefits manifest themselves. The PFS in the PD-L1—negative group does not seem very impressive in and of itself, but the OS [benefit] is present in that population.
What are some of the biggest challenges that future research will need to address?
The biggest challenge that needs to be worked out is finding more meaningful biomarkers. Does TMB further stratify our patients? Is it going to be two things we are looking at, PD-L1 status and TMB—using an algorithm? Are there more than just those two biomarkers to look at? It’s going to be interesting, as we have combination chemoimmunotherapy with additional agents that may further enhance the immunogenicity of these tumors. We are really in the infancy of immunotherapy in NSCLC; it's only been about 7 years since we even heard the words PD-1/PD-L1 inhibition. We are continuing to learn, and it's happening at a very fast rate. Every 6 months we have multiple new trials to sort through.