Letter to the FDA from a Mentor Employee
June 22, 2006
Food and Drug Administration
Center for Devices and Radiologic Health
Mr. Carl DeMarco, R.Ph., J.D.
Office of Device Evaluation
301 594 2022 134
Round Profile Gel Mammary Implants
Premarket Approval Application
Preclinical Chemical and Mechanical Testing
201 Mentor Drive
Santa Barbara, CA 93111
Research and Development Laboratory Services Document Control Program
1) Diffusion Testing Validation
The gel bleed or diffusion testing for low molecular weight siloxanes representing Round Gel Mammary Implants is invalid. I discovered this discrepancy shortly after the submission of the first PMA deficiency response (August 2004) upon thorough review of the raw data and results. Xxxxxxxxx performed the testing and failed to inform me of these difficulties. Consequently this information was concealed from me until I confronted her with the issues. Had I been aware of the problems I would never have submitted the work to the FDA since it could not be defended. The two major issues with the test method were the signal-to-background ratios (~0.1) and now recovery for the duration of the study. This was communicated with upper management Xxxxxxxxxxxxxxxxxxxxxx and also shared with other senior management personnel. I was eliminated from any further participation with FDA to discuss and provide an appropriate interpretation of the results. Subsequently the results were defended by Xxxxxxxxxxxxxxxxxxxx and no acknowledgement of the inability to validate the submission data was communicated. Eventually alternate methodology was developed, reviewed and approved by me (~1 year) however this also was not communicated since it would directly contradict previous submission data. The report for the new method is located in 1 of 2 file cabinets in Document Control labeled Gel Mammary Implants. This report outlines the new and distinct test method that was developed and validated. It describes the manner in which to present data using statistical analysis and includes recovery results for the 90 day duration of the diffusion experiment. The previous method could not be validated for recovery since after approximately 7 days the target analytes disappeared. Accordingly comparison of the old method results to the new method results will reveal the the former yielded erroneous data. Work ultimately returned to my oversight upon the reassignment of Xxxxxxxxxxxx. However no further communication was permitted with FDA during the RPG PMA review..
2) Explant Semivolatile Extractable Testing
Chemical testing of explanted implants was undertaken during the preparation of the deficiency response directed by Xxxxxxxxxxx. Again I was eliminated from participation due to my objection of the preliminary experimental design. Results showed that explanted device gel exhibited exceedingly large quantities of low molecular weight siloxanes compared to devices that had not been implanted. This report is located in 1 of 2 file cabinets in Document Control labeled Gel Mammary Implants. It was added as an addendum to another project conducted at SwRI (Southwest Research Institute) pertaining to total platinum analysis of shell and filler components in gel mammary implants. The immediate interpretation was in vivo biodegradation. Xxxxxxxxxxxxx responsibility for damage control. These results were not submitted to FDA in the deficiency response. No further studies have been undertaken to address the issues. Xxxxxxxxx were reassigned. I have separated from Mentor. Project status unknown.
3) Explant Mechanical Testing
Mechanical testing of explanted implants was undertaken during the preparation of the deficiency response directed by Xxxxxxxxxxxx. Results indicated that a significant reduction of some mechanical properties ha d occurred compared to devices that had not been implanted. These results were misinterpreted in a manner to conceal the relative change compared to control devices such that only absolute changes were reported during implantation. This presentation of the data yielded the apparent result of minimal degradation. No further testing was conducted to determine the origin of the discrepancies between control and explant test results. The original version of the report including control device data from PPQ manufacturing support and presenting percent retention of mechanical properties, in some instances showing a 50 % loss, is an electronic copy retained in Xxxxxxx Mentor directory.
4) Device Projection Fatigue Testing
Mechanical testing of device lifetime from fatigue has been underway since the deficiency submission. The influence of device projection on fatigue lifetime was investigated. Results showed that the device projection and lifetime were inversely proportional. In other words, as the device projection increased the estimated lifetime decreased. Both smooth and textured high profile devices yielded fatigue data statistically unique from moderate profile devices with a corresponding shorter lifetime. This has not been communicated with the FDA. Since I was the only scientist aware of this and I have separated from Mentor I don't know if this will ever be revealed or reported. The raw data for this work in on my desk in a folder label RPG Addendum I or it may also be in Xxxxxxxxxx office area since she was conducting the experiments. Coincidentally the same outcome for CPG was encountered after the PMA submission also indicating the higher profile devices have an apparent shorter lifetime. I'm certain this has also been concealed from FDA.
5) Platinum Valence
Chemical test data exists that indicates platinum is present in the shell with valence Pt (II). This is far more toxic than platinum valence Pt (0) which is the chemical species currently used for potential toxicological exposure. This information has not been communicated with the FDA. The work was conducted by Xxxxxxxxxx who was informed not to publish that information. I believe the results for gel showing Pt(0) were being prepared as a publication at the time of my separation from Mentor. The interim report showing the Pt(II) valence in shelll components is on my desk in a folder labeled Platinum. In addition Xxxxxxxxx has a copy in her office area since was a participant in this project.
It should be noted that much of this new data was generated after the PMA submission and that Xxxxxxxx would not support further updates since it may influence the decision of FDA to approve the RPG product. Much of Mentor personnel have been terminated after the divestiture of Urology. Only xxxxxxxx are still employed in R&D in Santa Barbara. Xxxxx is now part of Coloplast. Xxxxxx has retired. Accordingly it is unlikely if any of this information will ever be shared with the FDA. It is my understanding the laboratory is slated for complete shutdown in the next 90 days.
(former Mentor Employee)