Lung Cancer Survival Data May Help Merck’s Keytruda Catch Up

Top-line data from KEYNOTE-010 suggest opportunity for Keytruda to broaden its reach to a wider population with lower expression of the PD-L1 biomarker.

Reassuring evidence of an overall survival benefit for Merck & Co. Inc.’s PD-1 inhibitor Keytruda in non-small cell lung cancer could help the drug secure a better label that is more competitive with its closest rival, Bristol-Myers Squibb Co.’s programmed death 1 inhibitor Opdivo.

Merck announced the positive outcome for Keytruda (pembrolizumab) vs. docetaxel in the Phase II/III KEYNOTE-010 study in second-line NSCLC on Oct. 26; full results will be presented at a scientific meeting. In the trial, the drug demonstrated an overall survival benefit compared to docetaxel for patients at two cut-off levels for expression of the PD-L1 biomarker, ≥1% and ≥50%. Details about the magnitude of the benefit were not released.

The company plans to submit the data to the US FDA in late 2015 and to regulatory agencies in the EU in early 2016.

The OS benefit is satisfying but was expected, so it is “unlikely to change many forecasts of Keytruda uptake in the marketplace,” BioMedTracker analyst Robert Jeng commented. “In the end, this should lead to a fairly comfortable transition for Keytruda to full approval. We await quantitative details from the study and the eventual label,” he said.

The drug recently secured accelerated approval for second-line treatment of squamous and non-squamous NSCLC with high expression of the PD-L1 biomarker (50% or more, as measured by a companion diagnostic from Dako Corp.) (Also see "Can Merck’s Keytruda Start Change In Lung Cancer Market?" - Pink Sheet, 12 Oct, 2015.). Clearance was based on response rate data from the Phase I KEYNOTE-001 study.

In contrast, Bristol’s PD-1 inhibitor Opdivo (nivolumab) was approved in October for non-squamous NSCLC regardless of PD-L1 expression, based on results from a pivotal study that included overall survival data for the drug compared to chemotherapy (Also see "Opdivo Back On Par With Keytruda In Lung Cancer" - Pink Sheet, 9 Oct, 2015.). The drug had already been approved in squamous cell NSCLC since March 2015 and included in cancer compendia treatment guidelines for non-squamous NSCLC since July.

Leerink Partners analyst Seamus Fernandez commented in an Oct. 26 note that it’s challenging to determine the impact of the KEYNOTE-010 results without hazard ratios and overall survival data to compare side by side and that Bristol’s “broad” second-line label is a near-term advantage.

Satisfying The Survival Test

KEYNOTE-010 tested two doses of Keytruda, both given every three weeks – the approved 2 mg/kg dose and an investigational 10 mg/kg dose – to docetaxel in 1,034 previously treated NSCLC (squamous and non-squamous) patients with ≥1% expression of PD-L1. Both doses demonstrated an OS benefit for those with greater than 1% expression and 50% expression, Merck reported. Both doses also demonstrated a superior progression-free survival benefit in the ≥50% group, but the PFS improvement for Keytruda was not statistically significant for those with ≥1%.

Safety was consistent with previously reported data, the company reported.

The potential for expanding Keytruda’s label to include patients with lower PD-L1 expression will depend on the quantitative details and how different the survival benefit is between those at the 1% vs 50% cutoff levels, Jeng commented.

“This could be a huge boost to Keytruda, allowing it to target a larger population – perhaps even off-label in patients without biomarker data,” the analyst said.

The topline results appear consistent with results for Opdivo in Bristol’s pivotal CheckMate NSCLC studies, as widely expected, Evercore ISI analyst Mark Schoenebaum said in an Oct. 26 note, and should be enough for Merck and test-maker Dako “to expand Keytruda use (and the diagnostic assay) to patients with tumors at beyond the 50%+ PDL1 cutoff to the 1%+ cut-off,” he said.

Schoenebaum estimated that 60% of patients have more than 1% PD-L1 expression, as measured with the companion diagnostic.

The data could be incorporated into labeling in the first half of 2016, which “could help to level the playing field in 2nd line NSCLC,” Schoenebaum concluded.

Similarly, Morningstar Research analyst Damien Conover commented to “The Pink Sheet” that while the data “came in largely as expected,” the details will be helpful in determining what kind of label update Keytruda will likely get in lung cancer.

“I think it is likely Keytruda will get a label very similar to Opdivo in lung cancer for a relatively wide patient population and not just limited to PD-L1 positive patients,” Conover added.

Howard (Jack) West, medical director of thoracic oncology at the Swedish Cancer Institute in Seattle, said the data are “reassuring for proving what we were previously left to presume.” That is, he said, that Keytruda at 2 mg/kg IV every three weeks is effective in advanced NSCLC and that the benefits are greater in patients with higher PD-L1 expression.

“But these results merely follow the same pattern as nivolumab. I would like to see the actual data before drawing any further conclusions,” West added.

Shrugging Off PFS Result

Analysts were not very concerned about the lack of PFS benefit in the overall study population in KEYNOTE-010. A similar result was observed in trials of nivolumab in patients with lower PD-L1 expression.

Conover noted that tumors “respond to PD-1 drugs in a way that doesn’t make PFS as helpful a measure in determining a survival benefit.”

The fact that the PFS benefit was seen only in the highest-expressing patients is an interesting outcome, but it is unlikely to have any major impact on Keytruda uptake, Jeng said.

“It is becoming well-recognized that immunotherapies tend to provide an OS benefit, even without a PFS benefit due to the time required to mount an immune response. In this case, Keytruda only provided a PFS benefit in the highest-expressing patients which is the currently approved population anyway. Even if the label expands to include lower-expressing patients, we do not see this as a major issue,” Jeng concluded.