Betting on Biotech
Wall Street Analyst Who Sounded Alarm Over Prostate Cancer Drug Provenge In JNCI Settles with SEC Over Shorting Dendreon Stock
By Paul Goldberg
The lead author of a paper that argued that the prostate cancer immunotherapy Provenge endangered lives of cancer patients has entered a settlement agreement with the Securities and Exchange Commission.
The agreement states that Marie Huber, an analyst at a hedge fund, had prepared and anonymously circulated a paper about Provenge (sipuleucel-T) at a time when her put options in Dendreon Inc., the drug’s sponsor, were about to expire.
Huber subsequently expanded her arguments as the lead author of a 2012 paper in JNCI. Disclosure that accompanied that paper stated that Huber holds no position in the company, but stops short of disclosing that she had shorted the stock 18 months earlier. JNCI requires disclosure of relevant conflicts for a period of 36 months prior to submission.
In the agreement with SEC dated Nov. 27, Huber admitted no guilt, but agreed to pay a $25,000 fine and accepted a six-month suspension from participating in securities trading activities.
Her associate, Jess Jones, who played a supporting role in distribution of the document, also signed the agreement with SEC and was subjected to the same penalties.
More than anything, the story of Provenge, an autologous cellular immunotherapy for the treatment of asymptomatic or minimally symptomatic metastatic castration resistant prostate cancer, points to fluidity of the boundaries between the investment community, academia and advocacy groups.
Since March 29, 2007, the day the Provenge application was first considered by the FDA Cellular, Tissue and Gene Therapies Advisory Committee, these interests—in both the pro-Provenge and anti-Provenge camps—clashed ferociously (The Cancer Letter, April 13, 2007; April 27, 2007; May 4, 2007).
During the drug’s colorful history, the most unseemly behavior came from the pro-Provenge forces, a loose conglomeration of self-described patients and investors.
Many of these individuals—derided as “Dendreonites” and “Provengeans” by people who didn’t share their zeal—congregated on the Investor Village forum, whipping themselves into frenzy of enthusiasm for Provenge and vitriol toward skeptics.
At one point, death threats were made against doctors and scientists who advised FDA to delay approval until data from a randomized trial became available.
Money was collected on Investor Village to sue FDA, and unsuccessful efforts were made to seize computers from The Cancer Letter as part of discovery in that suit. An Ohio judge ultimately fined the plaintiff for using invalid subpoenas and engaging in behavior that was meant to harass this publication (The Cancer Letter, Dec. 9, 2010).
Dendreon wasn’t involved in these activities and was at times their target.
Now, Huber’s agreement with SEC and her failure to make disclosure to JNCI suggests that neither side was well-behaved. (This would be doubly disappointing if the hypothesis she puts forth in the JNCI paper is valid.)
And, after all that steam was expended, Provenge has failed to become the blockbuster drug its boosters envisioned.
Provenge hit the market in 2010 with the price tag of over $90,000, almost double what was then the going price of a cancer drug. Other pricey cancer therapies at that time clustered around $50,000 for a year of treatment.
Now, Provenge is getting competition from other therapies, primarily the Johnson & Johnson oral drug Zytiga (abiraterone acetate), which was approved for castration-resistant prostate cancer.
Today Dendreon’s stock is trading at around $3 a share. In late-April 2010, just before Provenge hit the market, the company’s shares were trading at over $54.
Now, the scientists who conducted the phase III clinical trial that led to Provenge’s approval are about to ask JNCI to retract the paper, arguing that the behavior described in the SEC settlement is inseparable from the paper ultimately published by the journal.
“The Huber paper published by the JNCI is a sad example of how financial interests have subverted the scientific process,” said Eric Small, co-director of the Urologic Cancer Service and director of urologic oncology research at the University of California San Francisco, is a co-author of a New England Journal of Medicine paper that Huber disputed in JNCI.
Philip Kantoff, the lead author on the NEJM paper that summarized the results of the Provenge pivotal trial, said the SEC disclosure provides grounds for retraction of Huber’s paper.
Kantoff, chair of the executive committee for clinical research and vice chair of Department of Medical Oncology, Solid Tumor Oncology at Dana-Farber Cancer Institute, said he isn’t surprised by this latest twist in the story of Provenge.
“Provenge has been one of the most polarizing therapies,” Kantoff said. “I think the reason for that goes deep into the history of immunotherapy, the believers vs. the non-believers. There was some amount of charlatanism, some amount of poor science that went into it, leading to a general disbelief that it actually would work eventually.
“The flip side of it were people who were strong believers in it who wanted it to be the answer to cancer.”
An Alternative Explanation
Though an advisory committee voted to approve Provenge in 2007, FDA decided that the company’s application was flawed and that a well-conducted phase III trial would be needed to support approval.
This delay so profoundly angered the Dendreonites that scientists who expressed skepticism received death threats. Howard Scher, chief of the Genitourinary Oncology Service at Memorial Sloan-Kettering Cancer Center, and Maha Hussain, associate director for clinical research at the University of Michigan Comprehensive Cancer Center, had to attend the 2007 annual meeting of the American Society of Clinical Oncology in the company of armed bodyguards.
Nonetheless, Dendreon completed a large randomized, controlled trial, IMPACT, announcing its results in a press release in April 2009, claiming a four-month survival advantage. FDA approved the agent on April 29, 2010 (The Cancer Letter, April 30, 2010) and NEJM published the results on July 29, 2010.
After reviewing the FDA approval documents, Huber concluded that the Provenge pivotal trial was fundamentally wrong.
Patients on the experimental arm lived longer than those on control arm because patients who got placebo received significantly fewer T-cells than Provenge arm patients. She believed that placebo patients received about one-third of their cells back, which caused them to die sooner.
In June 2010, she started work on a report titled “Provenge Ph III Trials – The Alternative Explanation of Survival Results.”
Huber, who at this writing is 35, was employed as an analyst at a hedge fund. She has a Bachelor’s degree in biochemistry and a Master’s degree in Bioscience Enterprise from Cambridge University. She is not a registered adviser and doesn’t hold any securities licenses, SEC documents state.
Her resumé, available on her website, states that between 2007 and 2011 Huber was an analyst with New York-based P. Schoenfeld Asset Management.
Documents suggest that Huber believed that the market would soon recognize the flaws that she believed marred the Provenge pivotal trial.
Between June 17 and July 12, 2010, Huber purchased $125,431 in July Dendreon put options and $110,627 in August put options.
“Huber also purchased put options in her mother’s account, and shared her analysis with friends and family who subsequently traded in Dendreon securities,” the agreement with SEC states.
According to the agreement, Huber didn’t receive approval for these trades, as required by her fund’s trading policies. Jones, who gambled a lower sum, similarly didn’t obtain approval from his fund.
The July put options were set to expire July 17, 2010, and had strike prices ranging from $10 to $30. All of the put options were “out-of-the-money,” and most of them had strike prices of $25 or less.
Dendreon common stock was selling in the low to mid $30s.
This was a gutsy move. Put options that aren’t exercised become worthless.
On June 30, 2010, the Centers for Medicare & Medicaid Services launched a national coverage analysis for Provenge, and requested public comment.
According to SEC documents, Huber encouraged her hedge fund to submit the report she prepared on Provenge to CMS. Though anyone can submit a public comment to a government agency, hedge funds generally do not.
“As the July 17, 2010, put option expiration date neared, respondents were concerned that HFA-A was not going to submit the report to the CMS website prior to the expiration of their put options,” the agreement states. “As a result, Respondents arranged to disseminate the Alternative Explanation on their own prior to option expiration.”
On July 12, 2010, Huber gave Jones a flash drive which contained documents relating to the alternative explanation, including copies of the alternative explanation, a distribution list of email addresses, and a version of the email text that Jones subsequently used to disseminate the report.
Documents state that on July 14, 2010, Jones created an email account using the name Jonathan White and Dendreon’s ticker symbol DNDN (jon.white.dndn@gmail.com) and sent emails attaching the alternative explanation to more than 450 email addresses from a distribution list that Huber had provided.
Most of the email recipients were affiliated with the medical and pharmaceutical industries, documents state.
The report that was attached to the “Jonathan White” emails, as well as Huber’s agreement with the SEC, are both posted on The Cancer Letter website.
The letter read:
Dear Colleague,
The document attached…was written by a group of scientists and physicians whose concern for their safety has forced them into hiding. In it they postulate a design flaw in the Provenge Ph III trials with potentially profound implications. Those who previously voiced legitimate scientific concerns regarding this drug had their lives threatened, were forced to employ body-guards and have been traumatized into silence.
Every dissenting voice is squashed. This fear extended to the FDA reviewers, who stated if it doesn’t get approved this time, there will be bloodshed. It is our constitutional right to express our opinions. If money and power can scare dissenters into silence, it is a sad day indeed for our nation and for humankind.
I call upon you to read this argument and make your own independent, critical assessment of its merits. If you see the merit of the concerns it voices, I call upon you to express those views to the FDA and CMS (Leslye Fitterman, PHD; Leslye.fitterman3@cms.hhs.gov) who have been trusted with the power of protecting the American public.
In my personal opinion (and that of select esteemed colleagues) that a legitimate concern has been raised, which is that the immune cells that are explicitly removed from placebo patients in the Provenge trials could have significantly compromised these patients and their ability to fight their cancer. This possibility must be explored as an alternative explanation for these trial results, because if it is right, it implies that Provenge treatment is harmful to patients because of all the immune cells that are lost during this treatment, and not prolonging life at all!
If any of you reach the same independent assessment of this piece as I do, it is our moral obligation to have a voices heard and demand this matter is investigated. We must stand up against those that wish to use the power of the sword to threaten legitimate scientific discourse and concern for patient safety. We cannot allow the big money invested in this drug to feed on the fear and desperation of cancer patients and their families to co-opt their voice to silence those very people that are trying to protect them.
Sincerely,
A concerned physician, scientist and citizen.
P.S. Scientific progress since 1999, when the FDA agreed to the design of these trials, has significantly increased our understanding of immune aging. Now that we know that the aged immune system cannot replace lost cells in the way that the youthful immune system can, we should identify the possible mistakes of our earlier ignorance. We infected thousands of people with HIV and hepatitis C through blood infusions before we discovered that these are blood-bourn pathogens. We gave thalidomide to thousands of pregnant women before we understood that this was causing birth defects. We used epo to drive hemoglobin levels to unhealthy levels until we learned that this is harmful. We make mistakes, and scientific progress reveals those mistakes. The sooner we rectify earlier mistakes, the sooner we curtail the unintentional harm we are causing.
Stock Price Drops, But Not Enough
On July 14, 2010, Dendreon shares closed at $33.99 on volume of 4,042,300. On July 15, 2010, after the “Jonathan White” emails were sent, Dendreon shares fell 7.2 percent intraday ($31.54) and closed down 4.5 percent ($32.45).
According to the agreement, trading volume on July 15 was 9,084,700, nearly double the average volume for the three trading days before and after July 15. That day, Huber sold 376 July put option contracts with strike prices of $27 and $30 for total proceeds of $2,841.
However, Huber and Jones “suffered significant trading losses because the vast majority of their put option contracts remained unsold or unexercised because they were so far ‘out-of-the-money,’” the agreement states.
According to documents, on July 14, Huber told her boss at her hedge fund that there had been a leak of the alternative explanation, and that she didn’t know who leaked it. Since the document was now in public domain, she urged her boss to submit the document to CMS as public comment.
The fund’s outside counsel submitted the alternative explanation to CMS on behalf of an unidentified client, and CMS posted the report on its website as part of the public comments. It’s legal for a hedge fund to submit comments to CMS. However, such actions are unusual. An anonymous comment from a hedge fund is all the more unusual.
While it’s appropriate for analysts to peruse publicly available data, such as FDA releases, SEC states that “the text of the July 14 ‘Jonathan White’ emails omitted to state material facts.
“The emails stated that the Alternative Explanation was ‘written by a group of scientists and physicians’ and was signed ‘a concerned physician, scientist and citizen.’ These statements were materially misleading because the respondents were hedge fund analysts who held Dendreon put option contracts that were about to expire. These facts were material because investors would have considered the identity, motive, and financial self-interest of respondents important to assessing the report and any decision to buy or sell the securities of Dendreon.”
From “Jonathan White” to JNCI
Contacted by The Cancer Letter, Huber responded with a statement, which was forwarded through her attorneys:
“I have agreed to a settlement with the SEC and I look forward to putting this issue behind me. I stand by the rigor of my scientific analysis and I shall continue to follow-up on the concerns that I expressed with regard to Provenge in the article published in JNCI in 2012.
“My primary motivation has always been the well-being of patients and the safety and effectiveness of this treatment. The accuracy of my research and conclusions are not an issue in the SEC settlement; neither my findings on Provenge submitted to CMS in July 2010, nor the concerns set forth by my co-authors and me in the JNCI article were contested by the SEC.
“Since first submitting the article to the JNCI in 2011, I have had no financial interest in the fate of Provenge. With this settlement complete, I look forward to moving forward with my life and career.”
The JNCI paper gave Huber’s argument something it lacked: scientific oomph. By the time she submitted the paper, she was no longer employed by the hedge fund.
She was working with three academic colleagues: Chris Parker, senior lecturer and honorary consultant in clinical oncology and prostate cancer translational research at the Institute of Cancer Research and the Royal Marsden, UK; Peter Iversen, professor of urology at Rigshospitalet, in Copenhagen; and Laura Haynes, then of the Trudeau Institute, of Saranac Lake, N.Y.
“I’m aware of the development, which of course is very disturbing,” Iversen said in an email to The Cancer Letter. “I am a co-author on the manuscript above. Apparently the first author, Marie Huber, have had an extremely unfortunate potential conflict of interest, which was unknown to me. However, I stand completely behind the arguments and concerns expressed, and questions asked, in the article. The content of the article remains an accurate reflection of my personal concerns about the IMPACT trial.”
“It’s very peculiar how academics got into this discussion along with someone who was working in a hedge fund,” Kantoff said to The Cancer Letter. “When that whole hypothesis came out, I asked a lot of expert immunologists and transplant physicians whether there is any potential validity to the contention that leukapheresis could be harmful, and I couldn’t find amongst a bunch of experts that there was anything that could potentially be harmful from it. I think this is purely conjecture not based on sound data.”
Revisiting the COI Disclosures
The JNCI instructions to authors state:
“Report all sources of revenue paid (or promised to be paid) directly to you or your institution on your behalf over the 36 months prior to submission of the work. This should include all monies from sources with relevance to the submitted work, not just monies from the entity that sponsored the research. Please note that your interactions with the sponsor of the work that are outside the submitted work should also be listed here. If there is any question, it is usually better to disclose a relationship than not to do so.”
In a separate section, authors are urged to “report other relationships or activities that readers could perceive to have influenced or that give the appearance of potentially influencing, what you wrote in the submitted work.”
In the disclosure form, a copy of which was obtained by The Cancer Letter, Huber stated that she has no conflicts to report, and added the following comment:
“I was previously employed by a registered investment advisor (until March 2011). Neither I, nor my former employer, has (or will have prior to publication), any financial interest in securities, either long or short, which may be influenced by the publication of this manuscript.”
On her website, www.theprovengetrials.org, Huber wrote that after leaving the hedge fund she continued to pursue the Provenge controversy.
Huber wrote that “neither I, nor my former employer, has any financial interest in the fate of Provenge.”
Huber’s description of her evolving interest in Provenge offers a glimpse at the manner in which analysts approach scientific inquiry:
“Unfortunately there was very little evidence publicly available to support the mechanism proposed by Dendreon. Provenge didn’t appear to affect disease progression and there was no published evidence of tumor killing. With most drugs I had previously researched, my comfort in their efficacy was based on studies, which provided evidence for the molecular and cellular mechanisms behind the outcomes being tested in trials. Because of this, the available evidence of Provenge’s efficacy came entirely from survival results of its phase III trials.
“A complicated history of changing endpoints and enrollment criteria, had left the trials with several design flaws. But the FDA had approved these changes, and since most of the flaws would probably result in relatively small biases, they needed to be weighed against the robust statistical significance of the 4.1 month survival benefit the trials had shown. I knew, however, that the FDA had access to significantly more data than the public, and so, when the FDA approved Provenge on April 29, 2010, I was fascinated to discover what data and analyses had been available to them to provide insights into the drug’s mechanism.
“Typically within 2-6 months after the FDA approves a new therapy, it posts a large set of its internal documents related to that approval on its website. And so it was that on June 9, I was able to begin scanning through these, and came across the confusing age data, which sparked my curiosity to dig deeper. Within days, the alternative explanation for the trial results became apparent, and in the weeks and months that followed, I spoke (under confidentiality agreements) with dozens of immunologists, immune aging experts, urologists and oncologists to solicit their opinions and see if there was any flaw in its logic or evidence against its plausibility that would justify dismissing it. While I heard many proposals for such counter-arguments, none of these held up to scrutiny.
“The story of the remainder of 2010 is long and complex, because my employer, for many very good reasons which I understood, did not want me to express and defend my concerns in public while I was still employed by the firm. Thus, in January 2011, I quit my job (a great job, with great healthcare benefits and a boss who was, and continues to be, a friend) in order to pursue the publication of the paper detailing the alternative explanation.”
It’s not publicly known how SEC became aware of the put options Huber purchased in 2010 or how it came to examine her role in writing the email blast on alternative explanations of the Provenge clinical trial. However, the JNCI paper was likely an invitation to scrutiny.
Huber’s agreement with SEC describes the paper as “a version of the Alternative Explanation,” which was “subsequently published in the Journal of the National Cancer Institute.”
The agreement states that the findings do not address Huber’s hypothesis.
A Dilemma for JNCI
Usually, a journal’s remedy for an author’s failure to disclose a relevant conflict of interest is mild: a correction.
However, in this case, the authors of the NEJM paper summarizing the IMPACT trial told The Cancer Letter that they intend to seek retraction of the Huber paper.
Charles Drake, associate professor immunology, urology and oncology and Johns Hopkins Kimmel Cancer Center, said the Huber paper should be retracted.
“I think the motivation was insincere,” Drake said.
The paper should be retracted because of the now acknowledged connection between the anonymous email and the article in a peer-reviewed journal.
The link between the email and the JNCI paper was immediately obvious. “Everybody either got it or passed it around,” he said.
“And when the article came out, that email, which seemed a little bit suspicious, had actually led to a paper. And the question is, who had actually written the email, who had written the article?”
Kantoff agrees.
“To say, ‘Oops, she didn’t disclose; she should disclose now,’ is, to me, not sufficient. I think it places into tremendous question the whole hypothesis that they bring forward that is made into an article that’s based on no data,” Kantoff said. “It’s based on ‘This is what we think.’
“It’s one thing to write a letter to the editor saying, ‘We think this,’ but they made this into an article, and it was co-authored by someone who is in a hedge fund, along with two academics, coming up with a non-scientific piece. And when I tried to rebut it, we were allowed to write a letter to the editor.
“I think JNCI handled it very poorly, and I would like to see retraction of the article.”
JNCI editors said they haven’t had the time to formulate a response.
“We are aware of the SEC settlement and are reviewing the relevant publications,” JNCI Editor-in-Chief Carmen Allegra said in an email to The Cancer Letter.
