FDA Approval: Osimertinib (Tagrisso) for Lung Cancer

A drug built to beat a specific escape route

FDA approved Tagrisso on November 13, 2015. The drug is osimertinib. Its file is new drug application 208065.

The approval letter states the indication exactly. Tagrisso was cleared for metastatic non-small cell lung cancer with an EGFR T790M mutation. The mutation had to be found by an FDA-approved test. And the disease had to have grown on or after an earlier EGFR-blocking drug.

That wording is doing a lot of work, so take it apart.

What EGFR is, and why the first drugs stop working

EGFR stands for epidermal growth factor receptor. It is a switch on the cell surface. Growth signals flip it on, and the cell divides.

In some lung adenocarcinomas the switch is faulty and stays on. NCI reports that EGFR and ALK changes dominate in tumors from people who never smoked. In one large series, EGFR changes turned up in 52 percent of those tumors.

Drugs that block that switch work well at first. Then, in many people, the tumor changes the switch's shape so the drug no longer fits. The commonest version of that change is called T790M.

Osimertinib was designed for the altered shape. That is why the 2015 indication needed a test showing T790M. It also needed an earlier EGFR drug to have failed first.

Approved on shrinkage, confirmed later

The approval letter carries a heading in capitals: ACCELERATED APPROVAL.

The letter says the indication was approved on tumor response rate and how long responses lasted. Continued approval could depend on proving real benefit in later trials.

Response rate counts how many tumors shrank by a set amount. It is not survival. It is a stand-in that regulators accept when a need is urgent and the signal is strong.

The confirmation came. FDA keeps a register of accelerated approvals that later proved benefit. Tagrisso is on it. Accelerated November 13, 2015. Verified March 30, 2017.

Our page on biomarker testing explains why a drug like this cannot be prescribed without a lab result first.

The label today is a different document

This is the trap in reading any approval story years later. Tagrisso in 2015 was a last-resort drug for one resistance mutation. It is not that now.

The current prescribing information lists five uses. Two of them involve EGFR exon 19 deletions or exon 21 L858R mutations found by an approved test. With those, the drug can be given after surgery, or for stage III disease that cannot be operated on. It is also a first-line treatment for metastatic disease with the same mutations, alone or with chemotherapy. And it is still approved for the original T790M group.

So the drug moved from the end of the line to the beginning of it. That is the real story, and a 2015 headline cannot carry it.

What it costs to take

The current label lists serious warnings, and they are not trivial.

First on the list is interstitial lung disease, also called pneumonitis. It means the lung tissue itself becomes inflamed and scarred. In the FLAURA2 study, using the drug with chemotherapy, it hit 3.3 percent of patients. Of those cases, 0.4 percent were fatal. In the LAURA study, after chemoradiation, 56 percent of treated patients had some lung inflammation, against 38 percent of those given placebo. Most of those cases were grade 1 or grade 2, and one was fatal.

Other warnings cover a heart rhythm change seen on an electrocardiogram, called QTc prolongation. Also cardiomyopathy, meaning a weakened heart muscle. Also keratitis, which affects the cornea of the eye. Also severe skin reactions, cutaneous vasculitis, and aplastic anemia.

Anyone taking it needs to report new or worsening breathlessness or cough promptly. That is the practical point buried in the pharmacology.

Who should ask about testing

Molecular testing is the gateway. Without it, none of this applies.

• Anyone newly diagnosed with non-small cell lung cancer should ask what molecular testing was ordered. Ask when the results are due.
• Never-smokers and light smokers with lung cancer have the highest chance of a targetable alteration.
• If an EGFR-targeted drug has stopped working, ask about repeat testing. It can be done on tissue or on a blood sample, and it may show why.
• Anyone with early-stage EGFR-mutated disease that has been removed by surgery should ask whether adjuvant treatment applies.

Our explainer on targeted therapy covers how these drugs differ from chemotherapy.

What this does not mean

• The indication in force is the current label, not the 2015 one. Check which applies.
• Accelerated approval on response rate is a bet that gets settled later. For this drug it was settled favorably. That is not automatic.
• These drugs work on a mutation, not on a cancer type. Without an FDA-approved test showing the alteration, the approval does not apply.
• Targeted drugs are not gentle. Lung inflammation on this drug can be fatal, which is why monitoring is part of the plan.
• Resistance is expected rather than surprising. Tumors evolve, and a drug that works can stop working.
• SEER, NCI's cancer surveillance program, puts five-year relative survival for lung and bronchus cancer at 29.5 percent overall. Once the cancer has reached distant sites, it is 10.5 percent. Both come from people diagnosed between 2016 and 2022, and both are group averages. They are not predictions, and they predate much of today's targeted therapy.

Sources

• FDA accelerated approval letter for TAGRISSO, NDA 208065 (PDF)
• FDA Drugs@FDA overview: TAGRISSO (osimertinib mesylate), NDA 208065
• FDA: Verified Clinical Benefit — Cancer Accelerated Approvals
• DailyMed: current TAGRISSO prescribing information (SPL)
• NCI PDQ: Non-Small Cell Lung Cancer Treatment (Health Professional Version)
• NCI SEER Cancer Stat Facts: Lung and Bronchus Cancer

How this article was prepared

An AI-assisted editorial system helped prepare this page. No named medical reviewer has reviewed it unless one is listed.

Cancer Explained is published by the National Cancer Information Foundation. It is not medical advice and does not suggest a test or treatment.

See an error, old source, or unclear wording? Tell us.