Gefitinib (Iressa) received FDA approval in 2003 as third-line treatment for metastatic non–small-cell lung cancer (mNSCLC). In phase 3 trials, the tyrosine kinase inhibitor shrunk tumors in only 11% of the efficacy population, but the change in that 11% was dramatic enough to warrant approval. When postmarketing studies showed no meaningful increase in response in a broader population, however, AstraZeneca voluntarily pulled gefitinib off the market. In 2012, the FDA withdrew its approval.
When gefitinib was approved, nobody knew why those 11% did so well. But in 2009, physicians at Massachusetts General Hospital and Harvard Medical School sequenced the EGFR genes in trial participants. Nearly all patients whose tumors had responded to gefitinib had mutations in the tyrosine kinase domain of the EGFR gene. No such mutations existed in nonresponders. This gave rise to the hypothesis that responders could be predicted if selected carefully.
AstraZeneca readied new clinical trials in patients with the relevant biomarkers. This time, response rates reached as high as 70%, and the duration of response exceeded that of the original studies. In July 2015, the FDA granted first-line approval to gefitinib in mNSCLC patients whose EGFR gene shows deletions on exon 19 or mutations on exon 21. The drug’s indication statement requires that these mutations be detected by an FDA-approved test.
On the heels of a disappointing phase 2 trial, AstraZeneca decided to end development of olaparib (Lynparza) as maintenance therapy for women with ovarian cancer. The placebo-controlled study of 265 patients whose cancer had relapsed produced a modest 3.6-month progression-free survival (PFS) advantage in the olaparib arm. AstraZeneca’s announcement in 2012 that it would terminate the program came even before investigators published their work in the New England Journal of Medicine.
But the investigators sensed that the outcomes weren’t telling the whole story. A subgroup of study participants with known BRCA gene mutations had shown a more promising response. Noting that their study was not designed to address differences among patients on the basis of BRCA status, the NEJM authors wrote that “there is a need to identify biomarkers to select patients for this therapy.”
Two years later, AstraZeneca retrospectively identified BRCA status for 97% of the women in the study. In 137 participants with BRCA mutations, the progression-free survival benefit was seven months, prompting AstraZeneca to request accelerated approval. Concerned about the sample size, however, an FDA advisory committee voted 11–2 against approval.
But in December 2014, the FDA recognized the underlying mechanism of disease and overruled the committee. It approved olaparib for women with germline BRCA mutations detected by an FDA-approved test. Continued approval may be contingent on the clinical benefit reported in ongoing phase 3 trials.
Payers are likely to watch these outcomes closely. At a cost of about $7,000 per month, olaparib exceeds conventional standards of cost-effectiveness, even when restricted to patients with BRCA mutations, according to a study presented at last year’s Society of Gynecologic Oncology annual meeting.
Ten months after the FDA approved ponatinib (Iclusig) as second-line treatment for some leukemias, the agency ordered Ariad Pharmaceuticals to stop sales and marketing. Real-world incidence of blood clots hit 27%, far higher than what the labeling showed, and fatal and serious events occurred in some patients just two weeks after starting therapy. “FDA cannot identify a dose level or exposure duration that is safe,” the agency warned in October 2013.
In the phase 2 trial that led to ponatinib’s approval, 54% of patients given the BCR-ABL inhibitor had achieved major cytogenic response (MCyR). But in a subset of patients—those with a BCR-ABL kinase domain mutation called T315i—MCyR reached 70%. This subgroup was no less at risk for vascular occlusion, but given its high response rate and the fact that the T315i mutation causes resistance to other kinase inhibitors (nilotinib [Tasigna], imatinib [Gleevec], and dasatinib [Sprycel]), the FDA knew that ponatinib addressed an unmet need.
Ultimately, the agency limited ponatinib’s indication to patients with a T315i mutation and approved a Risk Evaluation and Mitigation Strategy before sales resumed. Though the biomarker requirement cut ponatinib’s market to about 1,300 eligible patients in the United States, it may have saved the product and assured its availability to a handful of patients with few other options.
In returning to a market that has shriveled to a fraction of its original size, however, Ariad was forced to recoup R&D costs from a smaller group of patients. When ponatinib was first approved, its wholesale acquisition cost (WAC) was $9,580 for a 1-month supply. When it reintroduced the drug, Ariad increased the WAC price of ponatinib to $10,350, an 8% hike.