Conference Calendar

March 14-18 - American Association of Orthopaedic Medicine Annual Meeting 2017

March 14-December 31 - Canaccord Genuity Musculoskeletal Conference

Complete Calendar »

Earnings Calendar

No Upcoming Events Scheduled
Complete Calendar »

Read our research via:
email art

Weekly Email

rss art

RSS



app icon

iPhone

app store icon

Kindle



Orthopedic and Dental Industry News Complete Archive »

FzioMed's Oxiplex "Not Approvable" According to FDA Panel BY DAVID KRESSEL, JULY 16, 2008

On Tuesday, we attended the FDA's Orthopaedic & Rehabilitative Devices Panel, Oxiplex/SP was voted "Not Approvable" by a 5-2 vote, based largely on the sponsoring company's inability to definitively demonstrate product effectiveness. The two dissenters would have voted it "Approvable with Conditions".

Oxiplex is the lead product from FzioMed, a San Luis Obispo, CA-based biomaterials company. Oxiplex is used to prevent adhesions (peridural fibrosis), and is currently distributed by J&J and Medtronic outside of the US--it has been used in 100,000 surgeries to date. There is currently no FDA-approved product to treat spinal adhesions, though there are adhesion barriers that are approved for general surgery and ob-gyn indications. Oxiplex's indication would have been as a surgical adjuvant during posterior lumbar decompression surgery to improve patient outcomes by reducing post-operative pain. The pivotal study's primary efficacy endpoint was improvement in leg pain, as measured by the Lumbar Surgery Outcomes Questionnaire (LSOQ). Of note, the pilot study mentioned "reduction of adhesions" as the indication, but this was removed for the pivotal study.

FzioMed's presentation focused on a complicated "multivariate longitudinal” statistical analysis of the change in leg pain from baseline to six months by GEE (Generalized Estimating Equations)" which included ten covariates and five treatment-by-covariate two-way interactions. Their analysis showed that, in patients with more severe back pain before surgery, there was a statistically significant reduction in leg pain in the Oxiplex group.

The FDA's presentation focused on a series of simpler analyses, none of which demonstrated statistically significant reduction in pain scores. FDA's dataset was also different from FzioMed's because FDA included patients with results "out of window" for 6 month follow up (e.g. a patient who didn't visit their surgeon at 6 months after surgery, but 12 months after).

The panel sided with FDA and believed FzioMed's statistical method to be exploratory or post-hoc data mining. As one surgeon-panelist said, "I see complex data manipulation that I don't understand, and that worries me." Even the biostatisticians on the panel were skeptical of the validity of the analysis. The panel was able to unanimously concluded that Oxiplex:
-Was safe
-Addressed a significant clinical need
-Might be effective
-Did not prove effectiveness in its trial results

Disagreement surfaced over whether the product should be approved. Two panelists believed that the prospect of a potentially effective, positively harmless, product to treat an important problem with no other solution warranted approval.

Although many comments and questions from the panelists and FDA could have been judged as statistical nitpicking, the most interesting issue to arise was the fact that nobody, not even the company, had any clue as to how or why Oxiplex worked. FzioMed had a couple of unexplored hypotheses, but nothing else. Although concerning on its own, this also seriously undermined FzioMed's position regarding subgroup analysis, because if there is no understanding of the mechanism of action, there's no reason to trust post hoc analysis suggesting the product is more effective in a subgroup.

In the highly likely event FDA follows the panel's recommendation, it appears FzioMed will have to decide if it is interested in conducting a new clinical trial to prove efficacy in the subgroup of patients with worse baseline pain.

Email this to a colleague: