ESMO 2016: Spectacular data in NSCLC, but many questions remain

Advances in lung cancer treatment were probably the most anticipated and widely discussed topic at the 2016 European Society for Medical Oncology (ESMO) Congress in Copenhagen. Of especially high interest was the Presidential Symposium on Sunday 9 October, which led to chaotic scenes at the entrance to the main conference room - so big was the crowd of people wanting to attend the session, several overflow rooms had to be hastily arranged. The data that led to this exceptional situation were possibly the most important presented at an ESMO Congress.

Immunotherapies are at the centre of attention and are described with many superlatives like “the new cornerstone” of cancer therapy, “an earthquake” or “a game-changer.” If chemotherapy in the 1940s was the first wave of therapies, and targeted therapies starting in the 1990s were the second, then immunotherapies are the third wave and they are expected to be revolutionary. This expectation also explains the excitement when it came time to hearing the new data.

But as this is a relatively new area, many questions and uncertainties still need to be addressed and explored. Researchers are well aware of these, and first answers and directions were explored at ESMO 2016.

Biomarkers for patient selection

First, how should patients eligible for immunotherapy be selected? Can research identify promising biomarkers for patient selection or exclusion to determine the patients who are most likely to benefit from new kinds of treatment like PD-1 (programmed cell death protein 1) inhibitors?

PD-1/PD-L1 inhibitors belong to a class of cancer immunotherapies referred to as checkpoint inhibitors.  They work by inhibiting one of the mechanisms of resistance cancer cells develop to evade the human immune defence.

At the centre of attention were the data from KEYNOTE-024,¹ the premier phase 3 trial of first-line Keytruda (pembrolizumab, MSD) in patients with non-small cell lung cancer (NSCLC) and high PD-L1 expression (defined as expression in at least 50% of tumour cells). Pembrolizumab demonstrated superior progression-free survival (PFS) and overall survival (OS) over platinum-based chemotherapy in these pre-selected patients. KEYNOTE-024 met its primary endpoint, improving PFS by 4.3 months (10.3 months versus 6.0 months). The secondary endpoint of OS was also significantly prolonged. Along with the lower rate of treatment-related adverse events, these data point into a future where pembrolizumab may become the new standard of care in first-line therapy for patients with a high PD-L1-expression. “These data will completely change the management of patients with advanced NSCLC,” presenter Professor Martin Reck concluded.

Earlier in 2016, both PD-1 inhibitors Keytruda and Opdivo (nivolumab, Bristol-Myers Squibb) gained approval in second-line NSCLC, and the race for first-line started. In contrast to the celebrated data for pembrolizumab, nivolumab was not able to present convincing first-line data.

CheckMate-026², a study with chemotherapy-naïve patients with Stage IV or recurrent NSCLC compared nivolumab with a platinum-based regimen. Unlike in KEYNOTE-024 (≥50%), patients in CheckMate-026 were only required to show a PD-L1 expression ≥1% in their tumours. But the trial did not meet its primary endpoint in selected patients whose tumours showed a PD-L1 expression of ≥5%. PFS was 4.2 months with nivolumab compared with 5.9 months with chemotherapy. The OS could not be improved, and overall response rate was even lower.

An obvious way to try to explain the divergent results from these two phase 3 trials would be the different thresholds of PD-L1 expression used for patient selection. But it does not seem to be the PD-L1 threshold difference that can account for the lack of advantage in PFS for nivolumab, as a subgroup analysis of the CheckMate-026 data showed that even patients with PD-L1 levels of ≥ 50% did not show significant improvements in PFS or OS.

As CheckMate-026 presenter Dr. Mark A. Socinski put it: “If we are going to replace it [chemotherapy in NSCLC first-line therapy] with immunotherapy, we need to be confident that we are identifying the patients who will derive greater benefit.”

In his discussion of the results, Professor Johan Vansteenkiste stated, “More research is needed about how to use the PD-L1 biomarker to select patients for treatment with nivolumab.”

With Tecentriq (atezolizumab, Genentech/Roche), the first PD-L1 inhibitor (in contrast to the PD-1 inhibitors discussed above) presented phase 3 data in NSCLC from the OAK³ trial in the Presidential Symposium. A preliminary analysis of data demonstrated an improvement in OS (13.8 months versus 9.6 months) in patients treated with atezolizumab versus patients treated with docetaxel. This analysis was conducted independently from PD-L1 expression levels (including PD-L1 <1%). The OS benefit was seen regardless of PD-L1 expression levels.

Professor Martin Reck remarked: “We have a problem with using PD-L1 negativity as an exclusion factor for treatment.”

In conclusion, we have to realise that there is some interconnection between PD-L1 expression and the efficacy of PD-1/PD-L1 inhibitors, but it is not yet fully understood. PD-L1 expression might be an important parameter for treatment decision, but more knowledge and maybe other supporting biomarkers need to be explored in order to understand which patients will benefit the most and make an informed treatment decision accordingly.

Combination of immunotherapies

A second very important question is how the efficacy of immunotherapy can be improved. Potential combination partners can be either different immunotherapies, combinations with chemotherapies, or even with targeted therapies. A potential improvement in efficacy needs to be acquired without leading to an unacceptable level of adverse events.

Chemotherapy may have several immunologic effects (e.g., enhancing effector T-cell function) and it can induce PD-L1 expression on tumour cells, therefore representing a solid rationale as combination partner with immunotherapies.

Phase 1/2 KEYNOTE-021⁴ trial cohort G results imply that there is a benefit for patients adding pembrolizumab to chemotherapy in first-line therapy. This benefit is also likely to be proven in ongoing confirmatory phase 3 trials. The study met its primary overall response rate (ORR) endpoint, with a 26% improvement in the response rate (55% versus 29%). The downside is that Grade 3-4 adverse events increased with pembrolizumab (39% versus 26%). PFS curves separated at about one month, while the curves in the pembrolizumab monotherapy trial (KEYNOTE-026) separated only after three months. This can be seen as a suggestion that combinations of checkpoint inhibitors and chemotherapy will be the way forward.

Another promising approach might be the combination of different immunotherapies, e.g., a combination between the PD-1 inhibitor nivolumab and the CTLA-4 (cytotoxic T-lymphocyte-associated protein 4) Yervoy (ipilimumab, Bristol-Myers Squibb). The phase 3 trial, CheckMate-227, is currently enrolling PD-L1-positive patients who are assigned to receive either ipilimumab plus nivolumab or nivolumab alone or chemotherapy; PD-L1-negative patients will enter different study arms.

Even if positive results in combinations are already seen, and there will certainly be more to come, it also needs to be taken into consideration whether combination is really superior over drug sequencing. If the combination yields only minor benefits, it might not be worth giving away a valuable potential option for second-line by “wasting” it in a first-line combination therapy.

Treatment continuation of immunotherapies

A third challenge for which we do not yet have satisfactory answers is how long the immunotherapies should be continued in clinical practice.

With immunotherapy we are starting to see what is sometimes called the “long survival tail.” In the first couple of months immunotherapies seem not to yield an immediate effect, but after this time the survival curves start to separate from comparator treatment and eventually become flat and almost horizontal. This means that patients who are still alive after two years and still continuing on immunotherapies have a good chance of prolonged survival.

This leads to the question of whether immunotherapy actually be stopped after the 24-month mark without negative effects on patients’ health status. However, this is not solely a clinical discussion but is ultimately also a financial question as immunotherapies are quite expensive.

All of these questions and many more were discussed and debated at ESMO 2016, and we will see more answers to come over the next few years - immunotherapies will stay one of the hottest topics in NSCLC and oncology in general.

Meanwhile, the latest trial results will have an impact on first- and second-line treatments in NSCLC.  Pembrolizumab will start to be prescribed to first-line NSCLC patients with high PD-L1 expression. Patients with a lower expression will be offered a PD-1/PD-L1 inhibitor in second-line treatment, which will become an even more competitive space with atezolizumab entering beside the currently dominating nivolumab and pembrolizumab. The second-line market will be limited to some extent when patients already receive a PD-1 inhibitor in first-line. Something we will all continue to look at is whether nivolumab will be losing share in second-line after its disappointing trial results as a first-line treatment.

About the author

Markus Kaussen, Associate Director, Kantar Health. For more insights from Kantar visit the Kantar Health blog here.

References

¹ Reck M, Rodríguez-Abreu D, Robinson A, et al. Pembrolizumab or Chemotherapy in PD-L1–Positive Non–Small-Cell Lung Cancer. N Engl J Med. DOI: 10.1056/NEJMoa1606774

² Socinski MA, Creelan B, Horn L, et al.; “CheckMate 026: A Phase 3 Trial of Nivolumab vs Investigator's Choice (IC) of Platinum-Based Doublet Chemotherapy (PT-DC) as First-Line Therapy for Stage IV/Recurrent Programmed Death Ligand 1 (PD-L1) − Positive NSCLC;” Annals Oncol, 27 (s1)

³ Barlesi F, “Primary analysis from OAK, a randomized phase III study comparing atezolizumab with docetaxel in 2L/3L NSCLC,” ESMO Congress, Abstract LBA44_PR, Presidential Symposium 2, 2016

⁴ Langer C, Gaddgeel SM, Borghaei H, et al. ”Randomized, phase 2 study of carboplatin and pemetrexed with or without pembrolizumab as first-line therapy for advanced NSCLC: KEYNOTE-021 cohort G.” ESMO Congress, Abstract LBA46_PR, Presidential Symposium 2, 2016