Dainippon's Phase III Trial Shows Lurasidone Advantage Over Placebo, But Not Against Zyprexa

Like the newly-approved Saphris, lurasidone showed little impact on weight and other metabolic parameters in second Phase III PEARL study. The Japanese firm is going for a safety, not efficacy, entry point into the atypical class.

Dainippon Sumitomo Pharma Co. does have enough positive pivotal trials for its planned 2010 NDA filing for the atypical antipsychotic lurasidone, but the newly released data from one of those trials reveal a discrepancy in the efficacy results that could be a hurdle at FDA.

The Japan-based firm released the results of the Phase III PEARL 2 study Aug. 26, showing the novel schizophrenia drug at both 40 mg/day and 120 mg/day to be significantly more effective than placebo in treating schizophrenia, with effects on weight, total cholesterol and other lipid measurements similar to placebo.

It is the second positive Phase III trial for the drug, from a Phase III development program involving more than 2,000 patients, and Dainippon plans to submit an NDA in early 2010. Similarly positive results from PEARL (Program to Evaluate the Antipsychotic Response to Lurasidone) 1 were presented at the 2009 American Psychiatric Association meeting in May.

But although it is a placebo-controlled trial, and a greater effect was shown in that comparison, the PEARL 2 trial (unlike PEARL 1) also included an active-comparator arm that received olanzapine (Lilly's Zyprexa ) to establish assay sensitivity. Olanzapine outperformed lurasidone in the trial.

PEARL 2 was a double-blind, fixed-dose, placebo-controlled clinical trial with 478 schizophrenia patients randomized to be treated once-daily for six weeks with either lurasidone 40 mg or 120 mg, olanzapine 15 mg or placebo. [Editor's note: Dainippon emphasizes that the trial was not designed or powered for head-to-head comparison between lurasidone and olanzapine, and that while olanzapine did score numerically higher on the primary endpoint, it was not statistically significant.]

Both doses of lurasidone demonstrated significantly greater improvement versus placebo on the primary efficacy measure, a 30 percent or more improvement on the Positive and Negative Syndrome Score (PANSS) total score from baseline: 53 percent of patients on the 40 mg dose and 47 percent of patients on the 120 mg dose met that mark, versus 38 percent on placebo.

In comparison, 64 percent of the patients on olanzapine met the primary endpoint.

While the active comparator did better than lurasidone, it was a placebo-control trial design, and a placebo comparison is all that is needed for FDA approval.

This is a quandary that FDA (and an on-the-ball sponsor) worked through with Vanda's Fanapt (iloperidone). Iloperidone had some clinical trials that showed non-inferiority to an active comparator - although a significant improvement over placebo - and initially received a "not approvable" letter. With the help of a regulatory consultant, Vanda managed to secure approval in May based on the original data set - by making an argument that the apparent non-inferiority was not clinically significant ( (Also see "FDA's U-Turn On Fanapt: Comparative Assessments Must Have Clinical Meaning" - Pink Sheet, 1 Jul, 2009.) ).

Whether the Fanapt experience managed to establish a precedent that will aid Dainippon with lurasidone, or whether the comparative effectiveness question will again be a stumbling point, remains to be seen. It is also possible that the Dainippon trial design, by not including a non-inferiority comparison, manages to escape that problem.

Favorable side effect profile for weight gain, lipid effects

Dainippon is not seeking, however, to position lurasidone based upon its efficacy. Instead, the company is touting the safety profile for its product.

In PEARL 2, the incidence of clinically significant weight gain was 7.6 percent for the 40 mg dose of lurasidone, 4.2 percent for the 120 mg dose, and 7 percent for placebo. Lurasidone and placebo also showed similar changes in total cholesterol and other lipid measurements, and a discontinuation rate also on-par with placebo.

Those are all significant side effect problems associated with other schizophrenia drugs.

In contrast, 34.4 percent of the PEARL 2 patients on olanzapine experienced clinically significant weight gain.

Room for an eleventh?

Lurasidone's risk-benefit ratio appears to be similar to that of Schering Plough's newly-approved Saphris (asenapine), also an atypical antipsychotic shown to have status-quo efficacy but an edge in terms of metabolic safety parameters (Also see "Schering Likely To Play Up Saphris' Safety In Q4 Launch" - Pink Sheet, 14 Aug, 2009.).

But the "me-too" status of lurasidone should not be a prohibitive disadvantage against an atypical antipsychotic class that, while large, treats a hard-to-treat patient population that responds to therapies with unpredictable variability, leading to high switch rates and plenty of room for more drug options.

FDA's Psychopharmacologic Drugs Advisory Committee stressed this point when voting strongly in favor of Saphris just two weeks before the drug's approval ( (Also see "Panel Recommends Schering's Saphris; Safety Profile Could Provide Niche" - Pink Sheet, 3 Aug, 2009.), p. 10).

FDA agreed, approving Saphris as the tenth atypical, following other recent approvals of Fanapt and Johnson & Johnson's Invega Sustenna (paliperidone for injection) ( (Also see "Atypical Antipsychotic Market Snapshot: Making Way For Three New Drugs" - Pink Sheet, 24 Aug, 2009.), p. 19).

-Jamie Hammon ([email protected]) and Mary Jo Laffler ([email protected])