The discussions that took place in the executive offices of Memorial Sloan Kettering Cancer Center over the past week aren’t publicly known. Ditto conversations at top-tier medical journals and professional societies, which are assessing the aftermath of José Baselga’s systematic failure to disclose his conflicts of interest.
However, the extent of the disclosure problem created by Baselga, who stepped down as physician-in-chief and chief medical officer Sept. 13, can be assessed publicly—even in the comfort of one’s own office—thanks to a document Baselga himself created in order to explain his rationale for disclosing some competing interests while remaining mum on others.
If the document he created in response to questions from reporters from The New York Times and ProPublica is an indication, Baselga was guided by a highly unconventional interpretation of disclosure requirements and was inconsistent in applying even his own version of these standards.
As a result, all of Baselga’s 178 papers published between 2013 and now—while he was at MSK—may be tainted, experts say. Even in papers where Baselga has made a disclosure, the disclosure could be incomplete. Three other categories of papers, where he didn’t disclose, are: papers where he believed no disclosure was required, papers where the need for disclosure was, in his view, debatable, and papers where he admits an oops. According to Baselga’s analysis, 105 papers fall into these three categories.
Presentations made at meetings of professional societies are open to challenge, too.
Realizing that just one failure to disclose can ruin a career, most researchers consider it a safe practice to disclose everything, avoiding making their own judgments on relevance.
To explain his rationale to the Times and ProPublica, Baselga put together a list of the publications that cited him as an author between 2013 and 2018 and color-coded it, thus dividing his 178 publications into four categories:
“Publications where I disclosed
“Publications that represented basic laboratory or translational work
“Publications where disclosures are debatable. Most of them concerned exploratory biomarkers or reporting the results of early clinical trials with experimental agents for which a decision had been reached at the time of publication to not develop further. These publications had no clinical nor financial implications.
“Publications to be disclosed.”
A table below illustrates this breakdown:
disclosed | basic/translational | debatable | to be disclosed | |
2013 | 25 | 8 | 5 | 1 |
2014 | 21 | 6 | 5 | 2 |
2015 | 9 | 7 | 3 | 1 |
2016 | 7 | 19 | 5 | 4 |
2017 | 5 | 18 | 6 | 9 |
2018 | 5 | 4 | 2 | |
72 | 62 | 26 | 17 |
Baselga’s color-coded document—which is central to understanding what happened at MSK—was previously published by the Times and ProPublica, but, until now, hasn’t been analyzed publicly. The document is posted here.
The Cancer Letter asked three ethicists to review Baselga’s document:
“Dr. Baselga’s method of distinguishing which publications require disclosure is at least odd and appears to misunderstand the whole point of disclosure. Science develops incrementally, with each investigation building on the next. Scientists reading basic science reports or reports on biomarkers, which were not at this point targeted to be further developed, need to be alerted by disclosures to possible bias.
“Disclosures of financial ties can be helpful as one evaluates the report. No one is questioning the integrity of Dr. Baselga’s research, but the simple act of remembering one’s financial ties and disclosing them is a helpful step in keeping us all honest about our work.”
Rebecca Pentz
Professor of research ethics at Emory University Winship Cancer Institute
“As best I can tell, he seems to think unless his research involves a positive clinical trial, he need not disclose. I think. But, that standard is not what is expected by journals! Nor, if my speculation is right, does it make much sense in managing conflicts of interest which can arise in all manner of studies—basic, translational, negative and clinical.
“I think we should all be disclosing as required, but it is not all clear to me what difference disclosure makes. You still see many arrangements that raise an eyebrow or sometimes induce a yawn. I don’t think disclosure is the answer to managing conflicts of interest.”
Arthur Caplan
The Drs. William F. and Virginia Connolly Mitty Professor of Bioethics at New York University Langone Health and the founding director of the Division of Medical Ethics
“I looked this over. It is interesting to say the least with regard to the degree of parsing of the justification for non-disclosure that the investigator has undertaken—presumably after the fact. I think the principles which govern the reasons for why we ask investigators to disclose their potential and relevant COI prevent investigators from doing this kind of parsing. As well, my understanding of journal policies would prevent investigators from making these kinds of distinctions. That is, investigators should simply disclose their relationships, and the journals then can either, in turn, disclose these relationships within any articles they decide to publish or decide whether and which disclosures are relevant on their own and have those disclosed in the articles.
“As well, the very nature of the accepted definition of what a conflict of interest is, i.e., what a third party could perceive as a conflict that could cloud an investigator’s judgement, would prevent investigators from making these kinds of determinations of what’s relevant and what does not need disclosure. In fact, as it is in this case, it just seems too easy after the fact to judge the lack of disclosure coming as a result of clouded judgment—especially if the remuneration in some of these instances is as large as reported.”
Christopher Daugherty
Professor of medicine and chair of the University of Chicago Medicine Biological Sciences Division Institutional Review Board
The view that papers about basic laboratory and translational research do not require disclosure of competing interests is inconsistent with standards of most journals and professional societies, experts say. The same applies to studies focused on what Baselga describes as “exploratory biomarkers or reporting the results of early clinical trials with experimental agents for which a decision had been reached at the time of publication to not develop further.”
An explanation of the standards of the International Committee of Medical Journal Editors, which are used by many journals, reads:
“This section asks about your financial relationships with entities in the bio-medical arena that could be perceived to influence, or that give the appearance of potentially influencing, what you wrote in the submitted work.
“You should disclose interactions with ANY entity that could be considered broadly relevant to the work. For example, if your article is about testing an epidermal growth factor receptor (EGFR) antagonist in lung cancer, you should report all associations with entities pursuing diagnostic or therapeutic strategies in cancer in general, not just in the area of EGFR or lung cancer.
“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 work’s sponsor 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.
“For grants you have received for work outside the submitted work, you should disclose support ONLY from entities that could be perceived to be affected financially by the published work, such as drug companies, or foundations supported by entities that could be perceived to have a financial stake in the outcome. Public funding sources, such as government agencies, charitable foundations or academic institutions, need not be disclosed.
“For example, if a government agency sponsored a study in which you have been involved and drugs were provided by a pharmaceutical company, you need only list the pharmaceutical company.”
The standards of disclosure are now being applied to Baselga’s work retroactively.
“The American Association for Cancer Research has a longstanding track record of ensuring scientific integrity and ethics,” said an AACR spokesperson. “To this end, we regularly evaluate our policies and practices and are currently reviewing our conflict of interest guidelines. After this evaluation is completed, we will make any necessary revisions or updates.
“We are in review of Dr. Baselga’s documentation and will issue any necessary updates of the disclosures, as we have for other authors of articles in our scientific journals. This information will be visible to future readers of the journal articles in question.”
Baselga is a co-editor of Cancer Discovery, an AACR journal, and a past president of the professional society.
The American Society of Clinical Oncology is investigating, too.
“ASCO takes its disclosure and conflict of interest policies very seriously,” a spokesperson for the professional society said to The Cancer Letter. “Based on the specific concerns raised in the recent New York Times/ProPublica article, we have begun an internal review of Dr. Jose Baselga’s recent disclosures for annual meeting presentations as well as journal publications.
“Once ASCO staff have completed the internal review, ASCO volunteer leaders will be responsible for determining next steps, according to our established policies for both continuing education activities and journals. Once final decisions have been reached, they will be communicated to Dr. Baselga.”
Baselga is in communication with the New England Journal of Medicine, a spokeswoman for the journal said. “We’ve received updated forms from Dr. Baselga, and we have questions for him that must be answered for them to conform with our requirements. We await his answers,” said Jennifer Zeis, manager of communications and media relations for NEJM Group.
It just seems too easy after the fact to judge the lack of disclosure coming as a result of clouded judgment—especially if the remuneration in some of these instances is as large as reported.
Christopher Daugherty
JAMA Editor-in-Chief Howard Bauchner, said his journal is investigating. “JAMA is currently working with Dr. Baselga to clarify his conflict of interest statements in JAMA and the JAMA Network Journals,” Bauchner said in a statement.
The Lancet is looking into it as well. “We are investigating these claims, and take issues regarding conflicts of interests very seriously,” said Seil Collins, head of media and communications at The Lancet journals.
A spokesperson for Nature Research said its guidelines for authors are clear.
“Nature Research journals require authors to declare to the editors any competing financial and/or non-financial interests in relation to the work described that could be perceived by readers as inadvertently or deliberately influencing their presentation of the research,” the spokesperson said.
“The corresponding author is responsible for submitting a competing interests statement on behalf of all authors of the paper. This policy is clearly stated in our guide to authors, which also includes examples of financial and non-financial competing interests: http://www.nature.com/authors/policies/competing.html. Authors are encouraged to consider funding and employment while engaged in the research, as well as personal financial interests, such as stocks and shares, with organizations that may gain or lose financially through the publication of that specific paper.
“We believe that primary responsibility for ensuring that researchers’ conduct is appropriate lies with their employers, rather than with journal editors. However, should we become aware that the COI statement attached to a published paper is not compliant with our policies, we will look into the matter carefully and will update the literature when appropriate to ensure that the scientific record is accurate.”
