KEYNOTE-189 Updates Show Improved Outcomes for Metastatic Nonsquamous NSCLC

June 26, 2019
Nichole Tucker

In an interview with <em>Targeted Oncology, </em>Shirish Gadgeel, MBBS, of the Department of Medicine, Hematology/Oncology Division, University of Michigan, discussed recent study updates from the phase III KEYNOTE-189 trial and potential implications for the combination on clinical practice.

Shirish Gadgeel, MBBS

The phase III KEYNOTE-189 trial led to the FDA approval of the combination of pembrolizumab (Keytruda) plus chemotherapy for the treatment of patients with metastatic nonsquamous non—small cell lung cancer (NSCLC) withoutEGFRorALKalterations. During the 2019 ASCO Annual Meeting, updates to the study were presented, showing a longer follow-up and progression-free survival 2 (PFS2) findings.

Patients were randomized 2:1 in the phase III trial, receiving either pembrolizumab (n = 410) or placebo (n = 206) plus pemetrexed and carboplatin or cisplatin. Patients were given 4 cycles of treatment followed by pembrolizumab or placebo was with maintenance pemetrexed. In the occurrence of progressive disease, study subjects receiving chemotherapy were allowed to receive pembrolizumab.

The researchers gathered outcomes data and information on post-study anticancer therapy. PFS2 was determined from the time of randomization with first-line therapy until patients progressed on their next-line treatment. About 45% of the patients in the pembrolizumab plus chemotherapy arm received a second-line therapy while about 60% of the patients in the placebo plus chemotherapy arm received a subsequent therapy, including immunotherapy.

Pembrolizumab plus chemotherapy overall provided longer OS compared with chemotherapy alone in the overall population with a median OS of 22.0 months versus 10.7 months, respectively (hazard ratio [HR], 0.56; 95% CI, 0.45-0.70,P<.00001) after a median follow-up of 18.7 months. PFS was also significantly improved compared with placebo plus chemotherapy at 9.0 months versus 4.9 months, respectively (HR, 0.48; 95% CI, 0.40-0.58,P<.00001).

When looking at patients according to their PD-L1 tumor proportion score (TPS), among the patients with the higher PD-L1 expression (TPS >50%; n = 202), the results showed a an HR for OS of 0.59, HR for PFS of 0.36, and an HR for PFS2 of 0.47; additionally, the objective response rate (ORR) was 62.1% for patients with high PD-L1 expression who received pembrolizumab plus chemotherapy compared with 24.3% in patients who received chemotherapy alone. The group of patients with a TPS between 1% and 49% (n = 186) showed an HR for OS of 0.62, an HR for PFS of 0.51, and an HR for PFS2 of 0.59; the ORR in this subgroup was 49.2% with added pembrolizumab compared with 20.7% without. The group of patients with negative PD-L1 expression (TPS <1%; n = 190) showed an HR for OS of 0.52, HR for PFS of 0.63, and an HR for PFS2 of 0.46; the ORR was 32.3% versus 14.3% for patients who received the combination and chemotherapy alone, respectively.

The researchers confirmed that pembrolizumab can maximize outcomes when administered as part of first-line therapy in patients with metastatic nonsquamous NSCLC, with or without PD-L1 expression. Additionally, the investigators noted that safety and tolerability remained manageable with the combination even with longer follow-up.

In an interview withTargeted Oncology,Shirish Gadgeel, MBBS, of the Department of Medicine, Hematology/Oncology Division, University of Michigan, discussed recent study updates from the phase III KEYNOTE-189 trial and potential implications for the combination on clinical practice.

TARGETED ONCOLOGY: Can you provide background on the KEYNOTE-189 study?

Gadgeel: KEYNOTE-189 was a large, randomized phase III study evaluating the addition of pembrolizumab to chemotherapy of pemetrexed and a platinum in patients with metastatic nonsquamous NSCLC. In this trial, over 600 patients were randomized to either pembrolizumab with chemotherapy or chemotherapy alone.

The initial results of the study were first presented at AACR 2018 and simultaneously published in theNew England Journal of Medicine. And what the initial results showed was that there was a significant improvement in overall survival as well as progression-free survival with the addition of pembrolizumab to chemotherapy.

TARGETED ONCOLOGY: Can you explain the study updates that were presented at ASCO?

Gadgeel: At this year's ASCO, we provided an update with a longer follow-up. So, the initial follow-up was under a year when the initial results were published and presented. Now we have a median follow-up of almost 2 years. What we looked at was OS, PFS, and a new endpoint—a planned exploratory endpoint of PFS2. That is PFS on second-line therapy or death, in patients randomized on the study. Of course, we looked at if, with longer follow-up, we found any additional safety signals. And what we found with the longer follow-up is that the addition of pembrolizumab to chemotherapy did continue to show improvement in OS now with an HR of 0.56 that translated into median OS of 22 months with the addition of pembrolizumab to chemotherapy as compared to 10.9 months in patients who received chemotherapy alone. So, there was almost a doubling of the median survival. In addition, at 2 years we found that 45.5% of the patients were alive in the arm that received pembrolizumab plus chemotherapy whereas it was 29% for patients who received chemotherapy. So, there was almost a 16% improvement in survival at 2 years.

We also found improvement in PFS with the longer follow-up, now with an HR of 0.48 that translated into a PFS of 9 months with the addition of pembrolizumab to chemotherapy. And it was about 5 months with chemotherapy alone. So, again, a doubling of PFS.

But what was more interesting was that we found these improvements, both in OS and PFS, across all PD-L1 subsets including in patients whose tumor PD-L1 expression was 0. Similarly, the PFS2, which was the first time these data were presented for KEYNOTE-189, there was a significant improvement in PFS with an HR of 0.49.

And finally, with the longer follow-up, no additional safety signals were identified. So, the conclusion was, with longer median follow-up of almost 2 years, we continue to show improvement in OS, PFS, and now also in PFS2. But, importantly, these improvements were observed across all PD-L1 subsets including patients whose tumor PD-L1 expression was 0.

TARGETED ONCOLOGY: Can you explain the significance of the PFS2 endpoint?

Gadgeel: PFS2 is sort of an intermediate endpoint between PFS and OS. And one of the goals of this endpoint is to access the influence of first-line therapy on the therapeutic outcomes with the second line treatment. So, first-line therapy may negatively or positively impact second-line therapy. And what this data suggests is that there was continued improvement in PFS2 with this analysis.

TARGETED ONCOLOGY: Now with the approval of this combination and updated data, what do you think this says about how this combination therapy is impacting clinical practice?

Gadgeel: I think the greatest benefit of immunotherapy checkpoint inhibitors is in the treatment of NSCLC, either when used alone or when used in combination with chemotherapy, is that it tends to benefit not everybody, only some patients. But when they benefit, that benefit can be sustained.

This was actually best highlighted at this year’s ASCO meeting, not by the KEYNOTE-189 results but by the KEYNOTE-001 results. That was the first phase I study of single-agent pembrolizumab that showed that [in patients with NSCLC] whose tumor PD-L1 was high, 25% of the patients are alive at 5 years. When in the past, the 5-year survival was less than 2% in [patients with] advanced NSCLC. So, it's that sustained benefit that we see with checkpoint inhibitors that truly gets us excited.

Our hope is that a greater population of patients with NSCLC have an opportunity with the combination of chemotherapy and pembrolizumab, to potentially have that long-term benefit. So, hopefully, compared to single-agent pembrolizumab, a greater proportion of patients could potentially survive for a longer period of time—even up to 5 years. I think that is hopefully the greatest impact of KEYNOTE-189, but we will have to wait for a further update to truly understand what sort of patients and how many patients derive that kind of benefit.

TARGETED ONCOLOGY: What is your perspective on the use of tumor mutational burden (TMB) in lung cancer studies?

Gadgeel: So, there are several studies that have shown that patients whose TMB is high tend to benefit from checkpoint inhibitors, either single agent or combination. That is not just restricted to lung cancer but has been observed in other tumor types also. First, when it comes to KEYNOTE-189, [we are unsure of how] TMB [influences] the efficacy of pembrolizumab plus chemotherapy. That analysis is going to be conducted but we don't have those results as of yet. Even though we that patients with high-TMB tumors do benefit from checkpoint inhibitors—the magnitude of benefit is unclear. And how that compares to patients receiving chemotherapy with pembrolizumab is still unclear. My conclusion from the data we have to date is that I think we will use multiple biomarkers—PD-L1, TMB, and some others—to truly understand who benefits from single-agent checkpoints inhibitors, combination checkpoint inhibitors, or chemotherapy plus checkpoint inhibitors. But I think we need a lot of work to be done before we can have definite advances in this regard.

TARGETED ONCOLOGY: What are the unanswered questions or challenges with this combination?

Gadgeel: I think there are a few unanswered questions. One, is that the study was restricted to a performance status of 0 or 1. And in NSCLC, I would say anywhere from 15% to 20% of the patients have a performance status of 2 or higher [at the time of diagnosis]. Particularly for performance status 2, [we need to know if] this treatment is applicable and tolerable. I would also say that we need to enroll more elderly patients. The median age of lung cancer is 71 years [but] the median age enrolled in this study was in the range of about 65. So, this does mean that there are patients that we don't know how they will tolerate [the combination]. There is no indication to date that any of the immunotherapy studies, either a single agent or a combination, that elderly patients necessarily have an inferior outcome or more toxicity. But, I think that needs to be prospectively assessed to truly be able to say to an elderly patient that we can offer this treatment.

There are several questions as we see benefits with these treatments, but I would say that these 2 are the primary ones.

TARGETED ONCOLOGY: Are there any other important facts about KEYNOTE-189 that you would like to highlight?

Gadgeel: One thing I would say is that it is very important to recognize that KEYNOTE-189 excluded patients withEGFRmutation—positiveandALKtranslocation—positive NSCLCs. And I think, in this era, there is a concern with the KEYNOTE-189 results that physicians may get PD-L1 status on the tumor and just proceed with therapy [instead of waiting] for the mutation analysis. I think it is extremely important that we get the mutation profile of the tumor back before initiating therapy, because it would be wrong to say that chemotherapy plus pembrolizumab or pembrolizumab alone should be the therapy for all patients with NSCLC. So, we need to personalize the treatment based on the information we gather and for different patients, different treatments are [required]. I think it is very crucial that we pay attention to that aspect.


Gadgeel SM, Garassino MC, Esteban E, et al. KEYNOTE-189: Updated OS and progression after the next line of therapy (PFS2) with pembrolizumab (pembro) plus chemo with pemetrexed and platinum vs placebo plus chemo for metastatic nonsquamous NSCLC. J Clin Oncol. 2019;37(suppl; abstr 9013).