Clinical Trials Under Scrutiny

Over the last few years, as reports of liver failure began to surface among patients taking the antibiotic Ketek (telithromycin), Sanofi-Aventis's story about the integrity of its clinical data began to unravel. Although the FDA approved the drug, it asked Sanofi-Aventis (which was just Aventis at the time) to conduct additional safety and efficacy studies of the drug in several indications. The company agreed and outsourced the pivotal postmarketing study 3014 to a contract research organization (CRO). But when news of unscrupulous patient recruitment and significant underreporting of adverse events came to light, Sanofi-Aventis's finger pointing couldn't make the problem go away.

The issue is, when it comes to collecting—and using—questionable data, the company is far from alone. Over the last few years, there has been a growing number of instances of misconduct in clinical research. In fact, the number of allegations of research misconduct rose by almost 30 percent from 2005 to 2006, according to the Office of Research Integrity. And that may be just the tip of the iceberg: At least one study found that more than 40 percent of physician investigators (PIs) are aware of research misconduct but have not reported it to the appropriate authorities.

Much of this increase may be attributed to a change in the way companies conduct clinical trials. Though the industry traditionally contracted with academic medical centers to conduct sponsored clinical research, companies are now turning to physicians in solo or group practices. According to a recent study by the American Association of Medical Colleges, industry-sponsored research conducted in US academic medical centers decreased by 40 percent since 2004, in favor of community hospitals and independent physicians in private practice. Companies are making the shift because it gives them better access to patients and because private physicians and community hospitals have less infrastructure and overhead, which makes clinical research cheaper. But despite these benefits, trials conducted by these investigators often lack the auditing, monitoring, and other controls that are part of the academic medical center research environment, with their institutional review boards and clinical trial monitoring units. This makes sponsors more vulnerable to using questionable data and to charges of research misconduct—not to mention increased costs because they have to terminate and repeat those studies.

Companies began to recognize this issue in 2004, after the Chicago Tribune broke a story about a physician investigator in Kentucky who falsified the number of patients he enrolled in a clinical trial to increase his compensation. A study coordinator for the multi-site sponsored trial reported the PI to the sponsor after seeing the same lab results recorded for what were supposedly different human subjects on different dates.

In addition to falsifying data, physician investigators in private practice can also run into compliance issues associated with:

• research misconduct

• adherence to human subject protection

• billing problems

• disclosure of financial relationships and other related conflicts of interest.

At first glance, these issues may seem like the physician investigators' problems—after all, they are the ones engaged in the wrongdoing. But FDA's guidelines for Good Clinical Practice state that "ultimate responsibility for the quality and integrity of the trial data always resides with the sponsor." Associate US Attorney Jim Sheehan reaffirmed the accountability of sponsors for their clinical data during his talk at a compliance conference in September 2006 and noted that the Department of Justice (DOJ) intended to increase its scrutiny of fraud and abuse in clinical research.

Scrutiny by regulators and the media often leads to delays in development, lawsuits, expensive remediation, and sometimes irreparable damage to a sponsor's reputation. As a result, sponsoring organizations have a compelling interest in mitigating and protecting against noncompliance. This article provides insight into two particular areas—billing and conflicts of interest—because these are the areas most likely to be scrutinized in the near future.

Careful Billing

Regulatory scrutiny of clinical research intensified in 2000, following the adoption of the National Coverage Decision (NCD). The NCD defined new criteria for determining whether the government should reimburse medical procedures, items, and supplies given to Medicare beneficiaries enrolled as subjects in sponsored clinical trials. While the NCD may sound like a simple procedural change, it effectively meant that billing the federal government for non-covered services was considered fraud and abuse and was subject to severe sanctions under the False Claims Act.

The penalties for fraudulent billing are severe and not to be taken lightly. In a 2004 landmark settlement, for example, Rush University Medical Center in Chicago paid the United States $1 million after inappropriately obtaining Medicare reimbursement for cancer trial services not covered under the NCD. These are large fines for any organization. But these penalties are nearly impossible for private physician investigators and community hospitals to pay. In those cases, it's not unusual for physician investigators to turn to the sponsoring pharma company for help—as was the case in 2002, when the University of Alabama–Birmingham subcontracted lymphoma research from Biocryst Pharmaceuticals to an independent PI, who was subsequently convicted of falsifying research data. (See "Clinical Trials Gone Bad")

Indeed, while cases of fraud and abuse are often intentional, many physician investigators in private practice or community hospitals violate the NCD through lack of detailed knowledge. And when a revised NCD policy—with potentially even more stringent limitations—comes out in July 2007, that challenge may grow even greater.

Most physicians already find it difficult to distinguish between services provided for treatment and services provided for data collection. In particular, certain treatment areas, like cancer, are particularly confusing because most treatment modalities are experimental to begin with, and payers very often won't reimburse a therapy for unapproved uses. It is here that pharma companies can help physicians avoid costly billing errors by clearly identifying the services that are reimbursable under the NCD versus those that must be borne by the sponsoring organizations.

To address this liability, companies should make it a part of study protocol development to create a checklist of which services and procedures are needed strictly for the collection of data and which shouldn't be billed to payers. This checklist should be sent to the company's internal regulatory-affairs division when it reviews the study protocol for accuracy.

Sponsors should disseminate the checklist to physicians to ensure accurate and appropriate claim-form completion and submission to payers, especially with respect to the correct use of modifier codes identifying services or items provided in conjunction with a trial—an area of weakness within many provider organizations. A checklist is necessary even when investigators use billing companies that code and submit bills to payers based on medical records. In reality, many billing companies are less than accurate when it comes to modifier codes. They are also at the mercy of the physician's office—without a checklist, if the PI fails to note that a patient's visit was for a clinical trial, the billing company won't know how to obtain proper reimbursement.

Even if PIs or third-party billing companies assume the responsibility of following the NCD requirements, sponsoring organizations must be aware of and monitor adherence to such regulations. After all, while the responsibility to bill appropriately rests with PIs, at the end of the day, these private physicians don't have the expertise and knowledge that pharma companies do with respect to what services are part of the clinical trial. And when push comes to shove, companies are responsible for providing this education and oversight.

Conflict of Interest

Back in February 2006, the Journal of the American Medical Association published an article by 13 psychiatrists that found that pregnant women who stopped taking antidepressants during pregnancy fell into depressive relapses. It contradicted other credible studies that advised pregnant women not to take these drugs late in pregnancy because of the risk to the fetus.

While the results were widely publicized, one aspect didn't receive as much attention—at first. But a few months later, the news broke: The study authors all had financial ties to the pharma industry but had not disclosed them to JAMA.

Payment to physicians, and other conflicts of interest, is a ripe area of concern, particularly when it raises doubt over the integrity of research findings. Because many PIs are not as sophisticated as pharma companies about the issue of conflict of interest, companies should educate PIs on full disclosure, and its importance to research integrity.

A good way to do that is through the clinical trial agreement (CTA), the formal document describing the research protocol, schedule of events, and all contractual arrangements for PI compensation. Compensation, of course, must pass the fair market value test—regardless of whether the PI's participation is full- or part-time. When lump-sum payments to PIs are involved, it is essential that the CTA define unequivocally the allocation for overhead expenses, administrative or support services, and data collection and analysis, as opposed to medical services and supplies. Clinical trial agreements should also guide community hospitals and PIs to disclose any expenses for which the subject may be responsible once enrolled.

In addition, CTAs need to clarify the disposition of residual funds at the completion of the trial. If there is no clear definition of what to do with the money, PIs will be tempted to keep it for themselves—and perhaps even bill payers for the cost of conducting the clinical trial.

Finally, over the last few years, companies have begun developing ethics offices and other corporate-responsibility positions. These are appropriate channels for companies to use in offering guidance to PIs, but also for investigators who have questions about conflicts of interest. The ethics office is an important place for companies to compile and disseminate that information. And while companies are building these offices, they also need to encourage PIs to access them.

While there are many economic reasons for moving research away from academic centers, companies need to be aware that do so, they also lose the benefits of a well-monitored environment for clinical trials. They take on additional risks—risks that most companies are not managing at this point.

The intensified scrutiny of clinical trials by regulatory agencies, the severity of sanctions imposed for noncompliance, and the ongoing attention of the news media have indeed raised the level of compliance obligation among pharmaceutical companies and the independent physicians who engage in sponsored clinical research. At the same time, these external factors have introduced a new obligation on the part of sponsors and CROs to support, monitor, and, to the extent possible, sustain compliance efforts by participating physician investigators. Such collaborative efforts will lead to properly structured research programs that help clinicians enhance patient care, mitigate risks, and add value to their enterprises.

Timeline: Clinical Trials Gone Bad

Almost every day, the news media report fraud and abuse in the clinical research industry. The following examples show a history of infractions as a result of the lack of oversight by the PI's employer, the sponsors, the CROs, or the site-monitoring organization.

2002: The Department of Justice investigates Biocryst Pharmaceuticals and the University of Alabama-Birmingham for the falsification of clinical research data in a lymphoma study. The physician investigator, who is actually an independent subcontractor from the university, is convicted of the charges; however, the sponsor has to help the PI pay for the settlement because it failed to provide sufficient oversight to prevent the falsification of this data.

2004: After a study coordinator raises concerns about the quality of data supplied by a private physician in Kentucky, subsequent investigations reveal the PI had engaged in apparently fraudulent human-subject-enrollment practices to augment compensation.

2004: Rush University Medical Center agrees to pay $1 million to settle allegations that it billed the government for services provided in connection with clinical trials for which it had been reimbursed by the sponsor.

2005: Gary Kammer, MD, from Wake Forest University School of Medicine, resigns after he is discovered to have created "ghost subjects" in an NIH grant application for systemic lupus research.

2006: Falsification of study data leads to the resignation of Elizabeth Goodwin, PhD, from the University of Wisconsin, with subsequent data retraction and study termination at great costs to the study sponsor.

2006: A PI at the Cleveland Clinic allegedly fails to fully disclose that he received ongoing royalties from a sponsor of his clinical studies. Such a lapse violates the FDA guidance on financial disclosures by clinical investigators, designed to assure the integrity of study data.

Dinh Nguyen is director of compliance consulting with Sinaiko Healthcare Consulting. He can be reached at dinh@sinaikohc.com