Experts Consider the New Immunotherapy Paradigm in Advanced Lung Cancer

Get Permission

The fact that in the first-line setting, up to 30% of patients can directly get immunotherapy rather than chemotherapy is a major step forward.
— Suresh S. Ramalingam, MD

Tweet this quote

One immune checkpoint inhibitor has now moved to the front of the line for treating advanced non–small cell lung cancer (NSCLC). Based on pivotal studies presented at the 2016 European Society for Medical Oncology (ESMO) Congress, pembrolizumab (Keytruda) became a first-line option, and it is joined by two other checkpoint inhibitors as second-line options. How do experts view the data from the KEYNOTE-0241 and CheckMate 0262 trials, and how will clinicians apply these findings? The ASCO Post asked experts for their opinions.

Three antibodies targeting the programmed cell death protein 1 (PD-1) are now approved for advanced NSCLC—pembrolizumab, nivolumab (Opdivo), and atezolimumab (Tecentriq)—each of which demonstrated improved survival over docetaxel in patients progressing on platinum-based doublets. KEYNOTE-024 and CheckMate 026 evaluated pembrolizumab and nivolumab, respectively, as single agents in previously untreated patients. Only pembrolizumab met its primary endpoint (only treating patients with greater than or equal to 50% PD-L1 expression in the tumor), achieving not only a progression-free survival benefit (hazard ratio [HR] = 0.50; P < .001) but also an overall survival advantage (HR = 0.60; P = .005) over chemotherapy. The data led to the swift approval of pembrolizumab as a first-line option, whereas it was back to the drawing board for nivolumab.

First-Line Setting: ‘A Major Step Forward’

Suresh S. Ramalingam, MD, Professor of Hematology and Medical Oncology at Emory University and Deputy Director of the Winship Cancer Institute in Atlanta, commented: “The fact that in the first-line setting, up to 30% of patients can directly get immunotherapy rather than chemotherapy is a major step forward.”

If they have ≥ 50% PD-L1 expression, immunotherapy would be the treatment of choice. If it’s less than 50%, chemotherapy would be the recommended option.
— Karen Kelly, MD

Tweet this quote

Karen Kelly, MD, the Jennifer Rene Harmon Tegley and Elizabeth Erica Harmon Endowed Chair in Cancer Clinical Research and Professor of Medicine at the University of California Davis Comprehensive Cancer Center, agreed. She noted that the 30% of patients with programmed cell death ligand 1 (PD-L1) expression ≥ 50%, which is required for pembrolizumab’s use, is a doubling of the approximately 15% of patients with actionable mutations that permit targeted treatment.

“We have a whole new group of patients who can truly benefit from what I think is a well-tolerated, convenient therapy. It is really welcomed, especially for those patients without mutations, where we haven’t made much progress in front line for NSCLC,” she said.

Corey Langer, MD, Director of Thoracic Oncology and Professor of Medicine at the Hospital of the University of Pennsylvania, called first-line pembrolizumab “the ultimate game-changer in our therapeutic landscape,” one that will usher in a “complicated but also very exciting” shift in the treatment paradigm for advanced NSCLC.

Roy Herbst, MD, PhD, Ensign Professor of Medicine (Medical Oncology), Professor of Pharmacology, and Chief of Medical Oncology at Yale Cancer Center, New Haven, concurred. “Patients with targetable mutations should still get targeted treatment, but those who do not—whether squamous or nonsquamous histology—should get pembrolizumab,” he indicated.

Why Negative Results With Nivolumab?

Roy Herbst, MD, PhD

Roy Herbst, MD, PhD

In contrast, the experts admitted being a bit baffled by the failure of first-line nivolumab. “Why was nivolumab negative, with almost no benefit? That’s somewhat perplexing,” said Dr. Herbst. Dr. Langer echoed this sentiment: “I’m puzzled. I cannot really figure out why that trial was completely negative.” Dr. Ramalingam added, “I am not so surprised that pembrolizumab took the lead here. I am more surprised that nivolumab did not fare similarly to pembrolizumab.”

One question posed was whether subanalysis of the CheckMate 026 patients who mirrored the KEYNOTE-024 population—those with ≥ 50% PD-L1 expression—might derive an advantage. According to Dr. Ramalingam, such post-hoc analyses based on various PD-L1 expression thresholds did not reveal notable differences in outcomes.

This makes KEYNOTE-021 one of the first randomized, controlled trials of first-line therapy for advanced NSCLC in which median progression-free survival exceeded 1 year.
— Corey Langer, MD

Tweet this quote

Those interviewed agreed that the broader population in CheckMate 026—patients with expression ≥ 1% (≥ 5% for the primary analysis)—could have “diluted” the effect of nivolumab. Dr. Langer, however, noted that all patients expressed some degree of PD-L1 and none were PD-L1–negative. “You could argue that CheckMate 026 included patients who were probably less likely to benefit, but it was not a completely unselected population,” he pointed out.

Nevertheless, the two studies did have disproportionate numbers of patients with high PD-L1 expression as well as other imbalances, such as the use of radiotherapy (greater in CheckMate 026), and more patients in the nivolumab arm with liver metastases. Dr. Langer, however, cautioned against making too much of these observations. “It was a large trial, and I don’t think imbalances would have made that big a difference,” he maintained.

Is pembrolizumab simply a more effective antibody? “I don’t believe the drugs are that different,” Dr. Herbst said. Rather, he believes the explanation pertains to biomarker expression.

Dr. Langer, on the other hand, considered this possibility. “Until today, I didn’t believe there were differences in the drugs, but one now begins to wonder,” he commented. “Regardless, you cannot shine a laser pointer between those progression-free survival curves.”

“We have seen two studies with divergent results, and there are other studies in this space as well. As we learn more about these drugs, I think these issues will become clearer,” Dr. ­Ramalingam predicted.

All Patients to Be Tested

I will ask for the PD-L1 assay in essentially 100% of my patients, and patients will get pembrolizumab if they are eligible.
— David Spigel, MD

Tweet this quote

Until the presentation of date from the 2016 ESMO Congress, nivolumab was the favored anti–PD-1 agent in the second-line setting, since testing for PD-L1 expression was not required for its use, many oncologists agreed. “Now, we will test all patients, and the results will determine what we do next,” said David Spigel, MD, Director of the Lung Cancer Research Program at Sarah Cannon Research Institute, Nashville. “I will ask for the PD-L1 assay in essentially 100% of my patients, and patients will get pembrolizumab if they are eligible.”

“PD-L1 assessment must be part of our diagnostic workup, on a reflex and not reactive basis,” Dr. Langer agreed. “Patients cannot wait for these results.”

Targetable Mutations

All the experts agreed that patients with actionable mutations should continue to receive targeted agents in the first-line setting. “But at some point, those patients will come to a fork in the road and need to decide between chemotherapy and a PD-1/PD-L1 inhibitor,” Dr. Kelly said. “If they have ≥ 50% PD-L1 expression, immunotherapy would be the treatment of choice. If it’s less than 50%, chemotherapy would be the recommended option.”

For patients lacking mutations who receive first-line pembrolizumab and then progress, oncologists advise the patient to receive standard-of-care chemotherapy next or participate in a clinical trial,” Dr. Kelly indicated.

Second-Line Options

Immunotherapy options for the second-line setting now include nivolumab, pembrolizumab (with PD-L1 expression broadened to include ≥ 1% expressors), and atezolizumab, the third anti–PD-L1 antibody approved for previously treated patients.

In the phase III OAK study, which evaluated atezolizumab vs docetaxel as second- and third-line treatments, atezolizumab improved overall survival (HR = 0.73; P = .0003) regardless of PD-L1 expression status or levels.3 In patients with PD-L1 expression ≥ 1%, median survival was 15.7 months vs 10.3 months (HR = 0.74; P = .0102), but even PD-L1–negative patients derived benefit (HR = 0.75; P = .0215).

With nivolumab first on the scene (approved in March 2015), many—if not most—oncologists are familiar and comfortable with the drug and will continue to prescribe it, Dr. Ramalingam said. “However, the advantage to both pembrolizumab and atezolizumab is their every-3-week dosing, which may be more convenient to many patients and physicians,” he pointed out. “You save a visit every 6 weeks, and I think this will appeal to some physicians,” Dr. Spigel added.

Dr. Kelly said that since all newly diagnosed patients will now be tested for PD-L1 expression, “we will know the expression level for that patient, and then he or she will have to discuss with the physician which anti–PD-1/PD-L1 antibodies is right for him or her.”

However, Dr. Kelly noted: “Not all patients will have sufficient tissue to test for PD-L1,” and those patients would not get pembrolizumab (which requires the assay) but nivolumab or atezolizumab. “Patients who are untested can only receive the anti–PD-1/PD-L1 drug in the second-line setting.”

“The interesting question is whether patients who score ‘low’ for PD-L1 expression will still be offered nivolumab second line,” Dr. Spigel added. “The oncologist may say, ‘The patient is PD-L1–negative, so he should not receive nivolumab.’ I think that would be the wrong choice, but I think we could see some shift there in thinking they are not good candidates…. I think this would be a mistake.”

Dr. Spigel also predicted that nivolumab will probably remain a frequent second-line choice, but atezolizumb could surpass nivolumab because of its every-3-week dosing. Dr. Langer agreed and pointed out that the standard every-3-week dosing “dovetails” nicely with future combinations of agents. He also finds the atezolizumab data to be “robust” by showing a “fairly clear-cut survival benefit in the PD-L1–negative group and in never-smokers.”

With atezolizumab, you are also inhibiting B7.1. We are priming the immune system differently and activating the T cell differently. This may lead to differences in activity and also in the safety profile.
— Fabrice Barlesi, MD

Tweet this quote

According to the study’s principal investigator, Fabrice Barlesi, MD, of Aix-Marseille University and the Public Assistance Hospital in Marseilles, France, atezolizumab may act a bit differently from the other anti–PD-1 agents “because the interaction is not exactly the same as with the PD-1 inhibitors…. With atezolizumab, you are also inhibiting B7.1. We are priming the immune system differently and activating the T cell differently. This may lead to differences in activity and also in the safety profile.”

Based on potentially greater activity and convenient dosing, Dr. Langer said: “One could conjecture” that ­atezolizumab could potentially “displace” nivolumumab. Dr. Herbst concurred that pembrolizumab’s expanded indication in the second line for patients with any degree of PD-L1 expression could also mean a shift away from nivolumab, but he predicted that most oncologists will also consider their comfort level with the agents as well as reimbursement issues. “The main thing is that we want patients to get the best drug at the best time. They are all active,” concluded Dr. Langer.

Pembrolizumab Plus Chemotherapy

One could also foresee using pembrolizumab with chemotherapy in the first-line setting, based on the results of the phase II KEYNOTE-021 trial reported at the 2016 ESMO Congress by Dr. Langer.4 The study of pembrolizu­mab plus chemotherapy (carboplatin/pemetrexed) in nonsquamous patients of any PD-L1 status met its primary endpoint, with an overall response rate of 55% with pembrolizumab/chemotherapy vs 29% for chemotherapy alone (P = .0016). Responses were observed regardless of PD-L1 status, and median progression-free survival was 13.0 months vs 8.9 months, respectively (HR = 0.53; P = .0102).

“This makes KEYNOTE-021 one of the first randomized, controlled trials of first-line therapy for advanced NSCLC in which median progression-free survival exceeded 1 year,” Dr. Langer noted.

Dr. Kelly said she was impressed that patients benefited regardless of PD-L1 expression, “although those with the highest expression have higher response rates…. It still seems PD-L1 expression level matters.”

She could foresee the combination benefiting patients with less than a 50% expression of PD-L1, “but even for patients who get great benefit with pembrolizumab alone, the question is whether they could make that response more durable by adding chemotherapy,” she said. “We have not seen a lot of complete responses with these agents, and that’s something we will be looking toward, with the hope it will translate into even longer durations of response.”

Future Challenges and Opportunities

As important as pembrolizumab has become, its reign could be limited, experts interviewed by The ASCO Post predicted. “I think pembrolizumab will be the preferred first-line drug for a while, but I think there will be better strategies than single-agent checkpoint inhibitors,” Dr. Spigel offered. “I think ipilimumab [Yervoy] plus nivolumab will yield higher response rates than either single agent alone. A year from now, when the combination data read out, there could be a new first-line treatment overtaking pembrolizumab, and PD-L1 expression may not matter anymore.”

Dr. Ramalingam added that, in a phase Ib study, response rates to this combination exceeded 40%, and progression-free survival was “robust.” In addition to cytotoxic T-lymphocyte–associated protein 4 (CTLA-4), factors such as TIM-3, LAG-3, and OX40 can be co-targeted with a PD-1 inhibitor, he said.

“PD-1 and PD-L1 inhibitors in lung cancer have been approved at an incredible pace, and this pace will only be accelerated as more drugs come along. This space will change rapidly,” predicted Dr. Ramalingam.

There is also a big push to better understand which patients will respond and what factors are important beyond PD-L1 expression, Dr. Langer added. “We have to drill down to subpopulations,” he said. “We will be dividing this large population into separate niches for whom different treatment paradigms are going to emerge. These are expensive drugs. If they are not going to be better than chemotherapy alone, can we justify always using them?”

“We are happy that we can help more patients with this devastating disease, but we also recognize there is lots more work to be done,” Dr. Kelly commented. “As thrilled as we are at these results, we must remember there are lots of patients we haven’t yet helped, and we want to help them all.” ■

Disclosure: Dr. Ramalingam has served as an advisor to AstraZeneca, BMS, Merck, and Genentech. Drs. Kelly, Langer, Herbst, Spigel, and Barlesi reported no potential conflicts of interest.


1. Reck M, Rodriguez-Abreu D, Robinson A, et al: KEYNOTE-024: Pembrolizumab vs platinum-based chemotherapy as first-line therapy for advanced NSCLC with a PD-L1 tumor proportion score ≥ 50%. 2016 ESMO Congress. Abstract LBA8_PR. Presented October 10, 2016.

2. Socinski M, Creelan B, Horn L, et al: CheckMate 026: A phase 3 trial of nivolumab vs investigator’s choice of platinum-based doublet chemotherapy as first-line therapy for stage IV/recurrent programmed death ligand 1 (PD-L1)-positive NSCLC. 2016 ESMO Congress. Abstract LBA7_PR. Presented October 9, 2016.

3. Barlesi F, Park K, Ciardiello F, et al: Primary analysis from OAK, a randomized phase III study comparing atezolizumab with docetaxel in 2L/3L NSCLC. 2016 ESMO Congress. Abstract LBA44_PR. Presented October 9, 2016.

4. Langer CJ, Gadgeel S, Borghaei H, et al: Randomized phase 2study of carboplatin and pemetrexed with or without pembrolizumab as first-line therapy for advanced NSCLC: KEYNOTE-021 cohort G. 2016 ESMO Congress. LBA 46_PR. Presented September 9, 2016.