Keytruda NSCLC Approval Will Test PD-L1 Expression

Labeling is specific for those with PD-L1 expression as measured with a new companion diagnostic and does not reference data for PD-L1 negative patients.

FDA’s approval of Merck & Co. Inc.’s Keytruda (pembrolizumab) in non-small cell lung cancer marks a turning point in development for the family of programmed death inhibitors in that it is the first approval targeted for patients with PD-L1 expression, measured with a new companion diagnostic.

On Oct. 2, the agency granted accelerated approval for Keytruda for patients with advanced (metastatic) NSCLC whose disease has progressed after other treatments and with tumors that express PD-L1. The drug was approved for use with a companion diagnostic developed by Dako Corp. – the PD-L1 IHC 22C3 pharmDx test. In a statement about the approval, FDA noted that this is the “first test designed to detect PD-L1 expression in non-small cell lung tumors.”

Richard Pazdur, director of FDA’s Office of Hematology and Oncology Products, said in the statement that the approval “gives physicians the ability to target specific patients who may be most likely to benefit from this drug.”

What is perhaps the most interesting aspect of the label is the approval, and indeed requirement, of the first PD-L1 biomarker diagnostic, BioMedTracker analyst Robert Jeng told “The Pink Sheet” DAILY.

“There has been much confusion over the utility of PD-L1 biomarkers given the difficulty of their readout and inconsistency in their predictive ability. This approval solidifies the utility of this biomarker and could pave an easier path towards approvals in other indications for just biomarker-positive patients,” he said.

Merck had filed in April for "patients with advanced NSCLC whose disease has progressed on or after platinum-containing chemotherapy and an FDA-approved therapy for EGFR or ALK genomic tumor aberrations, if present." The drug had breakthrough status for that indication.

The filing, which was submitted in April and received priority review, was supported by results from the Phase Ib KEYNOTE-001 study in patients with at least 50% of tumor cells positive for PD-L1 (Also see "Keeping Track: AbbVie Is First With HCV Genotype 4 NDA, Merck Files Keytruda For NSCLC" - Pink Sheet, 27 Apr, 2015.).

However, there had been speculation about whether labeling would be inclusive of data regardless of PD-L1 expression, as there have been responses in PD-L1 negative patients in clinical trials. Merck exec Roger Perlmutter noted in July that it had submitted data for 100 patients who were PD-L1 negative and said there was a chance the results would be reflected in labeling to some extent (Also see "Opdivo Overtakes Keytruda, Fueled By Lung Cancer Use" - Pink Sheet, 3 Aug, 2015.).

The PD-L1 biomarker has been controversial in the clinical community, because status changes over time and those who are negative do benefit, just by a lesser degree compared to positive patients, and oncologists don’t want to see patients miss out on treatment (Also see "Building A Better Biomarker: PD-L1 Expression Under Spotlight At ASCO" - Pink Sheet, 8 Jun, 2015.).

Bristol-Myers Squibb Co.’s competing PD-1 inhibitor Opdivo (nivolumab) is already approved for lung cancer, for both PD-L1 positive and negative patients.

In the end, the Keytruda label does not reference PD-L1 negative NSCLC patients. Labeling reports results for a subgroup of 61 patients who had at least 50% tumor cells positive for PD-L1, showing a 41% response rate and that effects lasted from 2.1 and 9.1 months. “Among the 25 responding patients, 21 (84%) patients had ongoing responses at the final analysis,” the label states. There were no complete responses.

FDA’s letter noted that Keytruda has the potential to cause severe side effects that result from the immune system action of Keytruda.

In the 550 patients followed for safety, severe immune-mediated side effects occurred involving the lungs, colon and hormone-producing glands, FDA’s letter noted.

“Other uncommon immune-mediated side effects were rash and inflammation of blood vessels (vasculitis). Women who are pregnant or breastfeeding should not take Keytruda because it may cause harm to a developing fetus or newborn baby. Across clinical studies, a disorder in which the body's immune system attacks part of the peripheral nervous system (Guillain-Barre Syndrome) also occurred,” the letter warned.

The rate of severe adverse events was 38%, and included pleural effusion, pneumonia, dyspnea, pulmonary embolism, and pneumonitis. Long a concern in the lung, pneumonitis occurred in 19 (3.5%) of 550 patients with NSCLC, including Grade 2 (1.1%), 3 (1.3%), 4 (0.4%), or 5 (0.2%) cases, according to the label.

Labeling also noted occurrence of hyperthyroidism in 10% of patients and hypothyroidism in 6.9% of patients, and cited risk for developing type 1 diabetes. The drug was discontinued due to adverse reactions in 14% of patients.

The serious adverse event rate in melanoma, Keytruda’s initial indication, was 36%, with 9% of patients discontinuing due to adverse events.

Competitive Positioning

The approval gives Merck a chance to try and catch up with Opdivo.

Keytruda was the first PD-1 inhibitor approved in the US, with clearance for second-line metastatic melanoma in September 2014; Opdivo followed in the same indication in December that year. Both drugs are under review as monotherapies for first-line melanoma and Bristol’s combination of Opdivo with Yervoy (ipilimumab) was just approved for this indication (Also see "Bristol Hits Sweet Spot On Checkpoint Inhibitor Combination Pricing" - Pink Sheet, 1 Oct, 2015.).

The most valuable indication, however, is lung cancer. Opdivo was approved for squamous-type NSCLC in March and is being considered now by FDA for the more common non-squamous NSCLC, with a user fee date of Jan. 2, 2016.

Approval in lung cancer helped Opdivo to overtake Keytruda’s sales in the second quarter ($122m vs. $110m); before then, Merck had built a lead from its first-to-market status.

Bristol is also developing the Opdivo/Yervoy combination in lung cancer and has found a way to minimize side effects through a change to the dosing schedule (Also see "Bristol’s Opdivo/Yervoy Combo In Lung Cancer: Dosing Tweaks Improve Safety" - Pink Sheet, 8 Sep, 2015.).

Roche is expected to file its anti-PD-L1 atezolizumab in NSCLC in the first quarter, after releasing positive pivotal data at the European Cancer Congress in September (Also see "PD-1/L1 Update: Roche’s Atezolizumab Advances, Merck’s Keytruda Strikes In New Tumors" - Pink Sheet, 28 Sep, 2015.). AstraZeneca PLC is also readying its PD-L1 inhibitor durvalumab for submission in lung cancer.

Keytruda has a jump on Bristol in terms of approval for non-squamous NSCLC. However, the company is at a disadvantage in terms of supporting data. Keytruda’s approval is supported by response data, whereas Bristol has overall survival data for Opdivo in NSCLC. On the other hand, Keytruda’s response rates are strong and the clinical community views Keytruda and Opdivo as extremely similar drugs (Also see "Merck Asserts PD-1 inhibitor Keytruda’s Competitiveness In NSCLC" - Pink Sheet, 28 Apr, 2015.).

Bristol does have approval for the full range of PD-L1 expression in squamous type NSCLC. In Bristol’s CheckMate 057 study in non-squamous lung cancer, Opdivo demonstrated better survival compared to docetaxel in the overall study population. Those with low expression of PD-L1 did not have a survival benefit but they had long responses and a better safety profile. Hence, Bristol is expecting a broad label regardless of PD-L1 expression once again.

For Merck, the first look at overall survival data for NSCLC will come from the Phase II/III Keynote 010 study, which compares Keytruda to chemotherapy. That study is designed to end sometime around the end of the year, but an early termination is a possibility.

Merck has a broad development program for Keytruda beyond melanoma and lung cancer, with trials ongoing in 30 tumor types, and the company is testing Keytruda as part of many different combinations in partnership with other companies.