Report from the FDA Panel Meeting: Surgical Mesh for Transvaginal Repair of Pelvic Organ Prolapse

Last week, Dick Emmitt attended the FDA Obstetrics & Gynecology Advisory Panel Meeting convened regarding the use of surgical mesh for transvaginal repair of pelvic organ prolapse (“POP”). His report is below.

Venue:  Hilton Washington DC/North, Gaithersburg, MD.

Date:  February 12, 2019

FDA Stated Objective:  To obtain recommendations from the Panel to:

  1. Help guide the FDA’s evaluation of data from still ongoing 522 clinical studies that will be the basis of PMA’s for Restorelle (Coloplast), Uphold LITE (Boston Scientific) and Xenform (Boston Scientific)
  2. Help the FDA develop guidance to industry for future PMA study protocols for this indication.

The FDA also made it clear that the Panel would NOT be asked to opine on actual 522 study results that might be presented.

Attendees:  Approximately 200 (physicians, industry, device media, and consumers/patients)

Commercial Participants:  Boston Scientific and Coloplast

Strategic Background & Overview:

As a reminder, the 522 post-market studies were imposed on the industry as a result of a significant increase in transvaginal mesh MDR’s in the 2010-2012 period and the subsequent reclassification of these devices from Class II to Class III.  Of the dozen or so competitors on the market pre-reclassification, several never started a 522 study (including J&J and Bard) and two of the only four who started a study subsequently dropped out (American Medical Systems and A-Cell), thus leaving the two commercial participants in this Panel Meeting, Boston Scientific and Coloplast.   The remaining participants are now well past the one-year follow-up point and a high percentage (50% of BSC’s patients) are past the 3-year mark.

The actual deliberation of this Panel meeting was confusing and disjointed.   Despite  being initiated approximately 7 years ago, the presentation of actual 522 study results by the remaining two participants was apparently optional.  As a result, Boston Scientific provided a cursory review of its 522 study results (with only one-year follow-up) and Coloplast presentation included none of its 522 study results.  Including Boston Scientific, Coloplast, the FDA, and representative of several medical societies, the same  previously published clinical data and MDR summaries on old and no longer commercially available products were presented at least five different times.  Respecting the FDA’s very specific instruction to avoid opining on any new 522 study data, the Panel had only these presentations and their own personal experience on which to base its feedback.

Clinical/Regulatory Take-Homes:

The FDA’s “Questions for the Panel” were even more confusing, convoluted and repetitive than usual.  Panel members could never quite figure out whether a particular question was being asked in conjunction with evaluating the 522 data or establishing endpoints for future studies.  The lack of Uro/Gyn KOL panel members and constant opining by the non-surgeon Panel members also limited the productivity of the Panel discussion.  Despite all this, there were some interesting and important take-homes from this meeting:

  • Based on the data presented by Boston Scientific , the efficacy of the shared “control” group of the 522 studies represented by an AUGS Society data base of patients receiving transvaginal native tissue repairs (NTR’s) was much higher than expected (>90% compared to the general literature ranging from 60% to 80%) and more or less equivalent to the efficacy of the transvaginal repair arm at one year. The industry explanation (supported by virtually no presented data) was that the non-randomized structure of the trial resulted in younger/healthier patients being enrolled in the NTR control arm and older/sicker patients, not good candidates for an NTR, being enrolled in the mesh treatment arm.
  • Also based only on the BSC data at one year, somewhat offsetting the surprising efficacy of the NTR arm, the adverse event rate in the patients receiving a transvaginal mesh repair was very low with a very low incidence of mesh erosions/exposures that have been the source of so much bad publicity (BSC data only on Uphold LITE and Xenform).
  • Despite the representation (never in writing) that one-year data from the 522 studies would be the acceptable basis of a PMA, the FDA is going to require at least 2 years of data and probably another 3 years of post-market follow-up. The biggest concern is the long term adverse events of permanent implants and this concern was amplified by the personal testimony of several patients with particularly debilitating and tragic long term complications as a result of a transvaginal mesh repair.
  • Despite the FDA’s often-repeated assertion that mesh repairs need to offer a demonstrable improvement in efficacy to justify an increased risk compared to a native tissue repair, the Panel pushed back and repeatedly said that an improvement in efficacy is not required for a population that is not a good candidate for a NTR (due to lack of tissue, healing challenges, etc.). on It seems likely that the final labeling for transvaginal mesh (at least any mesh made from a non-resorbable material) will include some “second-line” language.
  • Going forward, it is very difficult to anticipate the FDA’s expectation for an IDE/PMA study structure to support a new mesh device. As was well-articulated by a Coloplast KOL adviser, RCT’s comparing mesh and NTR’s are just not feasible due to the lack of equipoise on the part of both physicians and patients.  Some type of OPC based on the AUGS NTR control arm data base seems most likely.
  • One point on which the Panel and the FDA were very aligned was the importance of “subjective” or “non-anatomic” outcomes, in other words, some type of Quality of Life (QoL) assessment. However, it was also clear that the FDA was positioning QoL as a potential “equalizer” for NTR citing “no difference” between NTR and mesh in a meta analysis of studies including those with superior anatomic results for mesh.