An Avastin Recommendation & Conflicts Of Interest

Earlier this month, the National Comprehensive Cancer Network, a non-profit group of oncologists whose guidance is closely followed by leading treatment centers, voted overwhelmingly in favor of maintaining its recommendation that Avastin should be used to treat breast cancer. The vote came shortly after an FDA panel voted 6-to-0 to revoke the breast cancer indication for Avastin.
The endorsement is important because oncologists will likely continue to use Avastin even if FDA commish Margaret Hamburg rescinds the breast cancer indication. Roche and its Genentech unit had appealed a decision last December by the agency to pull the indication for their best-selling med after new studies showed the med does not prolong overall survival in breast cancer patients or provide a sufficient benefit in slowing disease progression to outweigh significant risks. This prompted the unusual two-day hearing last month (back stories here and here).
However, 10 of the 33 members of the NCCN breast cancer panel members have ties to Roche or Genentech, either as advisory board members, speakers, consultants, expert witnesses or having received clinical research support. These connections are disclosed on the NCCN web site (look here). And 25 members of the panel participated in the recent vote to maintain the recommendation.
Specifically, the NCCN panel voted 24 in favor, 0 against and 1 abstention. The simple math suggests that at least one panel member – and possibly two – with ties to Roche voted to support the metastatic breast cancer recommendation. Perhaps more panel members with connections voted, although there is now way to know ascertain this since the NCCN press release does not specify who participated in the voting.
As we have noted previously, the NCCN endorsement is likely to be a boon for Roche, since treatment for breast cancer has typically generated about $1 billion or more in annual sales. Avastin rings registers – worldwide sales last year totaled about $6.8 billion and rose 9 percent, which meant this one drug accounted for 14 percent of total Roche sales. In other words, much is at stake.
Meanwhile, the stated NCCN policy conflicts of interest requires “disclosure of external relationships and recusal of NCCN Guidelines Panel Members with conflicting interests so that the integrity of the NCCN Guidelines is not compromised or diminished by conflicts or by the perception of conflicts,” according to the NCCN web site.
The policy also states that a panel member with a significant and direct or indirect relationship with “an external entity” that constitutes a conflict shall not participate in NCCN Guidelines Panel discussions, when the panel’s action on the topic under discussion “may advantage or disadvantage an external entity.” An exception is granted when requested by the panel chair “to participate for the purpose of providing or presenting information to the NCCN Guidelines Panel.”
More specifically, certain “direct relationships,” such as a panel member who is a beneficial owner of stock in an “external” entity or a director of such an organization” would be considered to have a de facto conflict. The policy also defines “direct relationships” as anyone “who receives compensation for services including, but not limited to, management or consulting services to the organization” (here is the policy).
So we asked NCCN whether this policy was followed for the recent breast cancer panel, given that the vote tally suggested otherwise. The spokeswoman repeatedly declined to discuss specifics and referred us back to the recent press release which, again, offers no information on the topic. In fact, she refused to answer whether NCCN has a recusal policy, even though this exists on the web site. “I’m only allowed to discuss what is in the press release,” she told us over and over.
We also reached out to the 10 panel members who have ties to Roche and Genentech. One responded. Antonio Wolff wrote us to confirm that “Genentech provides funding to Johns Hopkins University (where I am employed as School of Medicine faculty) to support research costs associated with an ongoing early phase clinical trial, and I am the site PI for that study. As for your specific question regarding my activities within NCCN, I will ask (you) to contact it directly as NCCN requires all panel members to adhere to its confidentiality policy.”
And so, an influential panel with ties to a drugmaker – which has a lot of sales on the line – voted to maintain a key recommendation. In this instance, NCCN panel members fully disclosed their ties to Roche, but is this sufficient? Supposedly, there is a reason NCCN has a disclosure and recusal policy, but in this instance, there would appear to have been a breach. If none occurred, the organization should be willing to discuss specifics and defend its policy. Yet NCCN refused to do so. What do you think?
Source: Pharmalot