Kim Discusses Multiple Immunotherapy Trials in PD-L1–Expressing NSCLC

Case-Based Roundtable Meetings Spotlight, Case-Based Roundtable Meetings Spotlight: January 1, 2022,
Pages: 4

During a Targeted Oncology Case-Based Roundtable event, Chul Kim, MD, MPH, assistant professor at Georgetown University, discussed immunotherapy options for patients with non–small cell lung cancer with PD-L1 expression.

Targeted OncologyTM: What therapies come to mind for a patient such as this?

KIM: Cemiplimab [Libtayo] has been recently approved by the FDA for treatment of advanced non–small cell lung cancer [NSCLC] with high PD-L1 expression.1 They define advanced as patients with locally advanced disease who are not candidates for surgical resection or chemoradiation, or who have stage IV. They used the 22C3 antibody [PD-L1 IHC 22C3 pharmDx kit] to define PD-L1 expression.2 The TPS needs to be at least 50% and you need to rule out other genomic alterations such as EGFR, ALK, or ROS1.

The approval was based on the EMPOWER-Lung 1 trial [NCT03088540].3 In this trial they included patients who were treatment naive with advanced NSCLC; either stage IIIB or IIIC, not eligible for definitive treatment or metastatic disease, PD-L1 at least 50%, and no other genomic alterations. They also included some patients who had clinically stable central nervous system metastases and controlled hepatitis [B or C] or HIV to reflect a real-world population, because a lot of these patients were excluded in other clinical trials.

As far as I know, a lot of the landmark trials, say pembrolizumab [Keytruda] in the KEYNOTE trials…excluded patients with hepatitis and HIV. This trial did not exclude those patients, but [overall] not a lot of those patients were enrolled in this trial. I think one of the fears was viral reactivation when these drugs were developed. So, KEYNOTE trials all excluded those viral treatments in NSCLC. But as we knew more about the safety and efficacy of immunotherapy in those patients with other infectious diseases and chronic illnesses, I think trials became more inclusive, including this trial.

How did investigators design the EMPOWER-Lung 1 trial?

The patients were randomized either to cemiplimab orchemotherapy.3 Then, at the time of progression, there was an option of adding chemotherapy to cemiplimab if the patient was assigned to arm A. If the patient was assigned to chemotherapy alone at the time progression, crossover was allowed in this trial.

The primary end points were overall survival [OS] and progression-free survival [PFS] and they included about 710 patients. What should be noted is that among these patients, about 560 patients had PD-L1 high [expression], at least 50%, because there were some cases where the PD-L1 expression scoring was not done according to the protocol. So they reexamined many of the patients, and then looked at 560 patients with a true PD-L1 high expression as a separate group.

If you look at the baseline characteristics, about 10% of patients had brain metastases and were clinically stable. About 15% to 18% of patients had locally advanced disease, so stage III disease, and then the rest of patients had metastatic disease.

What results were observed in this study?

The outcomes were of patients who were randomized, those 710 patients, including a minority of patients whose PD-L1 level was uncertain or less than 50%.3 If you look at the OS, the Kaplan-Meier curve [showed] the hazard ratio was favoring the cemiplimab, a hazard ratio of 0.68 [95% CI, 0.53-0.87; P = .002]. The benefit is seen also for the PFS [HR, 0.59; 95% CI, 0.49-0.72; P < .0001].

If you look at those patients with PD-L1 expression at least 50% after reexamination of the PD-L1 status, the hazard ratio for death was 0.57 [95% CI, 0.42-0.77; P = .0002]. So there was a 43% risk reduction in death. The median OS was not reached with the cemiplimab, and [was] 14.2 months with the chemotherapy. A similar PFS benefit was seen in this patient group [HR, 0.54; 95% CI, 0.43-0.68; P < .0001].

I should note that about 74% of patients had crossover from chemotherapy to cemiplimab at the time of progression, so there was a high rate of crossover. But despite that, the trial was able to show an OS benefit.

There were benefits seen in different subgroups, including those patients with brain metastases and in stage III vs IV.3 The objective response rate was 39% with cemiplimab and 20% with the chemotherapy [odds ratio, 2.53; 95% CI, 1.74-3.69; P < .0001]. Duration of response was longer with cemiplimab, 16.7 months vs 6.0 months with chemotherapy. Most of the patients had a response at the time of first scan at 2 months.

They looked at different PD-L1 subgroups, 90% or higher, [61% to 89%], 50% to 60%, [and less than 50% or unknown]. What they showed is if the patient had high PD-L1 expression, they had a better outcome. So if the PD-L1 expression is getting closer to 100%, the outcomes are very good in terms of the tumor dynamics and other results including OS and PFS.

There were no new safety signals observed in this clinical trial. We know that immunotherapy can cause various immune-related adverse events [AEs]. There was a higher chance of developing grade 3 or 4 AEs with chemotherapy compared with cemiplimab.

How has pembrolizumab been evaluated in the NSCLC setting?

We all are very familiar with the KEYNOTE-024 trial [NCT02142738]…one of the landmark trials in our field.4,5 This compared pembrolizumab with chemotherapy alone and there was also crossover built into this trial. The patients received about 2 years of treatment and then there was an option of rechallenge [for] the patients at the time of progression with a second course of pembrolizumab.

The baseline characteristics were pretty well balanced between the groups. There were 39 patients who were able to complete the 2 years of treatment with pembrolizumab and then about 12 patients received a second course of pembrolizumab at the time of progression.

A recent update was presented at the 2020 European Society for Medical Oncology [Congress] showing the OS benefits with pembrolizumab with a hazard ratio of 0.62 [95% CI, 0.48-0.81]. The median OS was 26.3 months with pembrolizumab [95% CI, 18.3-40.4] and 13.4 months with chemotherapy [95% CI, 9.4-18.3]. The median OS doubled with pembrolizumab compared with chemotherapy and so a clear survival signal was seen in this trial. If we look at the 5-year update of PFS, the hazard ratio is 0.50 favoring pembrolizumab [95% CI, 0.39-0.65].

In terms of objective response rate, pembrolizumab was associated with about a 46% objective response rate vs 31% with chemotherapy. Duration of response was durable with pembrolizumab [at a median of] 29.1 months vs 6.3 months with chemotherapy. In terms of the AEs, there were more grade 3 or higher AEs with chemotherapy compared with pembrolizumab; the rate was 53% with chemotherapy and 31% with pembrolizumab.

What other immune therapies could be considered for this patient?

Atezolizumab [Tecentriq] was approved as monotherapy in those patients with a high PD-L1 level based on the phase 3 IMpower110 trial [NCT02409342].6 In this trial, patients were randomized in a 1:1 fashion to atezolizumab vs chemotherapy. And atezolizumab was given every 3 weeks at 1200 mg until there was progression [of disease] or a loss of clinical benefit. The chemotherapy was given per histology and there was no crossover allowed in this trial. The baseline characteristics [were] comparable with [those of] prior studies.

OS was prolonged with atezolizumab in this clinical trial.6 The OS benefit was more prominent in those with a high PD-L1 expression and the PD-L1 expression here in this trial was defined by SP142 assay. In the cemiplimab and the pembrolizumab trials, they used a 22C3 antibody.1-3 Here they used a different antibody, SP142, and they not only looked at the tumor PD-L1 expression but also tumor-infiltrating immune cell PD-L1 score as well.6 The tumor PD-L1 score had to be at least 50% by SP142, and the immune cell expression had to be at least 10% by SP142 to be able to be defined as high PD-L1 expression. The hazard ratio for death was 0.59 favoring atezolizumab [95% CI, 0.40-0.89; P = .01]. The median OS was 20 months with atezolizumab vs 13 months with chemotherapy.

There is a higher concordance between 22C3 and SP263, and these assays look at tumor cells. SP142 looks at both tumor cells and tumor-infiltrating immune cells. So that may be one of the reasons why there’s some difference in general [because] SP142 has [different] analytic parameters compared with other platforms. But the bottom line is that with all the platforms, be it 22C3 or SP263, there was OS benefit seen with atezolizumab compared with chemotherapy. Then if you look at the AEs there was, of course, a higher chance of developing grade 3 or higher AEs with chemotherapy vs atezolizumab—there’s a different safety profile between those 2 treatments.

How do these therapies compare?

If you look at those 3 immunotherapy monotherapy agents—cemiplimab, atezolizumab, and pembrolizumab— there are some differences. Atezolizumab is approved for metastatic disease, so stage IV setting.6,7 Cemiplimab and pembrolizumab are approved for stage IV plus stage III if the patient cannot get surgery or definitive chemoradiation therapy.2,8

So that’s 1 difference between these drugs. Atezolizumab was approved as monotherapy only for high PD-L1 expressors; cemiplimab was the same thing, high PD-L1. Based on the KEYNOTE-042 [NCT02220894] data, we can use pembrolizumab for patients with PD-L1 expression [of] at least 1%.9

There are some differences in terms of the frequency of the dosing. Atezolizumab comes as [a regimen that can be given] every 2 weeks, every 3 weeks, or every 4 weeks.7 Cemiplimab is given every 3 weeks.2 Pembrolizumab has every-3-weeks [and every]- 6-weeks options.8 Across all these agents, we need to keep in mind that immunotherapy-related AEs can happen. Common ones are rash, diarrhea, hypothyroidism, immune-related pneumonitis, and myocarditis.2,7,8 So we need to keep an eye on our patients after we start immunotherapy.

Do you know why there is inconsistency in PD-L1 levels between the different trials?

I think that’s a really good point. Cemiplimab and atezolizumab have requirements for PD-L1 high levels, at least 50% for the tumor cells.2,7 Based on KEYNOTE-042, we can use pembrolizumab if the PD-L1 expression is at least 1%.8,9 However, if you look at the KEYNOTE-042 data between PD-L1 1% to 49%, there was no clear benefit with pembrolizumab over chemotherapy.9 In my clinical practice I tend to use chemotherapy plus immunotherapy in those with lower PD-L1 expression. But in patients with PD-L1 high levels we can use immunotherapy alone comfortably.

Then, in the cemiplimab data, there is evidence to suggest that the higher the PD-L1 expression is, the better the outcomes are.3 It’s a pristine biomarker, but there’s a tendency to have better outcomes in PD-L1 expression [if it] is close to 100%.

For pembrolizumab, do you tend to stop therapy at 2 years if patients are still responding? If they relapse, do you prefer to restart the pembrolizumab?

I tend to use [it for] at least 2 years and then I [talk] with the patient, and this is an area we don’t have a good answer for. If the patient has achieved close to a complete response, then I [tell] them we can either stop or continue beyond a 2-year time point. I think patient preference here plays a pretty big role because when we have no data to guide us, some people say, “Why would you rock the boat? Just continue what we are doing.” In my practice and my colleagues’ practice, we have several patients who [have been] on immunotherapy for 5, 6, 7 years. Many of them started in clinical trials of immunotherapy doing very well, [with a] complete response.

Some patients say that 2 years [on pembrolizumab] sounds like enough, and this treatment is getting tough with various AEs, so they want to stop. So we don’t know the exact answer [of when to stop treatment], and trials were conducted differently. With pembrolizumab, [in] the KEYNOTE trials, there was a hard stop at 2 years and across pretty much all pembrolizumab trials. But in other trials, [such as those with] atezolizumab, [treatment] was given until progression.6


1. FDA approves cemiplimab-rwlc for non-small cell lung cancer with high PD-L1 expression. FDA. February 22, 2021. Accessed December 5, 2021.

2. Libtayo. Prescribing information. Regeneron Pharmaceuticals, Inc; 2021. Accessed December 5, 2021.

3. Sezer A, Kilickap S, Gümüş M, et al. Cemiplimab monotherapy for first-line treatment of advanced non-small-cell lung cancer with PD-L1 of at least 50%: a multicentre, open-label, global, phase 3, randomised, controlled trial. Lancet. 2021;397(10274):592-604. doi:10.1016/S0140-6736(21)00228-2

4. Brahmer JR, Rodríguez-Abreu D, Robinson AG, et al. KEYNOTE-024 5-year OS update: first-line (1L) pembrolizumab (pembro) vs platinum-based chemotherapy (chemo) in patients (pts) with metastatic NSCLC and PD-L1 tumour proportion score (TPS) ≥50%. Ann Oncol. 2020;31(suppl 4):S1181-S1182.

5. Reck M, Rodríguez-Abreu D, Robinson AG, et al. Five-year outcomes with pembrolizumab vs chemotherapy for metastatic non-small-cell lung cancer with PD-L1 tumor proportion score ≥50. J Clin Oncol. 2021;39(21):2339-2349. doi:10.1200/JCO.21.00174

6. Herbst RS, Giaccone G, de Marinis F, et al. Atezolizumab for first-line treatment of PD-L1-selected patients with NSCLC. N Engl J Med. 2020;383(14):1328-1339. doi:10.1056/NEJMoa1917346

7. Tecentriq. Prescribing information. Genentech, Inc; 2021. December 5, 2021.

8. Keytruda. Prescribing information. Merck Sharp & Dohme Corp; 2021. Accessed December 5, 2021.

9. Mok TSK, Wu YL, Kudaba I, et al; KEYNOTE-042 Investigators. Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomised, open-label, controlled, phase 3 trial. Lancet. 2019;393(10183):1819-1830. doi:10.1016/S0140-6736(18)32409-7