Don't Rush to Judgment on Off-Label Drug Promotions

Maurie Markman, MD

Controversy over the FDA’s role in regulating off-label marketing practices of pharmaceutical companies continues to reverberate in the medical community. In a recent editorial, leading bioethicists cited a 2012 US Court of Appeals ruling that struck down the FDA’s ability to control the speech of a pharmaceutical company representative as long as the language was “truthful and nonmisleading.”¹

The editorialists disagreed with the decision in part because “it conflicts with the core values of the FDA because it restricts the FDA’s ability to promote public health by regulating communications concerning benefits and risks of medical products.”¹

Experience Contradicts Label

Although there certainly should be restraints on inaccurate or inappropriate promotion of medications, the question of whether pharmaceutical representatives should be required to adhere to the approved product label when discussing the use of antineoplastic agents is much more complex and nuanced. Is it truly in the public’s interest for a pharmaceutical company representative to be barred from discussing with physicians data from peer-reviewed literature that may help clinicians improve the administration of a drug?Consider, for example, the experience of the oncology community with pegylated liposomal doxorubicin in the management of recurrent and platinum-resistant ovarian cancer.2 The FDA approved this drug for this clinical indication based on data from an initial phase II trial and a subsequently conducted randomized phase III trial. In these studies, the initial drug dose delivered and approved was 50 mg/m² every 28 days.

However, it soon became apparent to all who administered the drug that this dose level produced excessive and unacceptable symptomatic toxicity such as mucositis, stomatitis, and Palmar-Plantar erythrodysesthesia (hand-foot syndrome) in a substantial percentage of patients. This was particularly concerning for a population often being treated with short-term palliative intent.

As a result, a number of investigators explored and subsequently published their experiences employing a lower dose of the agent (most often 40 mg/m2 every 28 days). Although there has not been a randomized trial comparing the approved versus the lower dosing levels, considerable data support the therapeutic equivalence of the regimens and the clinically meaningful superior toxicity profile of the lower-dose strategy (40 mg/m2 ).²

Nevertheless, since the dose in the original approval was never changed, either because such a request was not made by the manufacturer or because the FDA thought insufficient evidence existed to permit this modification, the regulatory rules governing commercial speech did not permit the company’s representative to discuss with treating physicians any dose level other than the more toxic 50 mg/m2 regimen.

Labels May Be Too Narrow

Instead of protecting the public, rules that deny pharmaceutical company representatives the ability to provide information that might improve the management of care for a patient with cancer because the material is not contained in the label represents a failure to promote public health through an open discussion of the benefits and risks of medical products.Another illustration of the sometimes strikingly narrow clinical indications for oncology drugs and the impact on a meaningful dialogue between representatives of the pharmaceutical industry and practicing clinicians concerns the initial regulatory approval of several PARP inhibitors for patients with previously treated epithelial ovarian cancer. Two novel agents, olaparib and rucaparib, gained regulatory approval for commercial sale based on phase II trial data demonstrating high objective response rates with relatively long times to subsequent disease progression.³

However, the label for olaparib initially stated that it should be prescribed after a patient has received at least 3 prior treatment regimens while the label for rucaparib states that it can be used after a minimum of 2 prior regimens. Considering the available preclinical and clinical data regarding PARP inhibitors in general and the specific trial results for these 2 clinical agents in ovarian cancer, was this difference in approved clinical indications necessary or even appropriate? Is it really in the public interest for antineoplastic drug approvals to be so specific and narrow?

Importantly, we should recognize that the drug label does not define the practice of medicine once a pharmaceutical agent is approved for commercial sale. Therefore, what public health mission is served and how is patient welfare enhanced by mandating that a pharmaceutical company representative follow an approved script from the FDA?

Another facet of this discussion involves requirements for a company and its representatives to include commentary regarding the relationship between a drug’s administration and diagnostic testing results, including the potential use of a specific molecular testing platform.

For example, at a time when there was widespread knowledge and substantial data to support the clinical utility of olaparib in patients with somatic BRCA mutations in addition to individuals with germline abnormalities, the initial approved indication on the drug label referenced only germline mutations.³ And, in the approval of rucaparib, a companion diagnostic was specifically referenced, but not the fact that widely accepted molecular testing platforms that meet Clinical Laboratory Improvement Amendments standards also could provide the required information regarding the presence of either a germline or somatic BRCA mutation.

Again, we must recognize that the FDA’s on-label approval does not equate with how clinical medicine must be practiced, but a strong argument can be made that the scientific authority and basic credibility of this vitally important regulatory agency is diminished in the eyes of both medical practitioners and the public when the agency focuses its efforts on areas that can hardly be classified as promoting public health.

Nothing stated in this commentary should be taken as support for inappropriate promotion of products without sufficient evidence of clinical utility. But there can be considerable value in the communication of data and experience beyond that included in the drug label, particularly in light of the rapidly escalating costs of antineoplastic agents and the major concern these expenses pose for our society.

References

1. . Hey SP, Cohen LG, Adashi EY, Kesselheim AS. Influence, integrity, and the FDA: an ethical framework. Science. 2017;357(6354):876-877. doi: 10.1126/science.aao1056.
2. Markman M. Serious ethical dilemma of single-agent pegylated liposomal doxorubicin employed as a control arm in ovarian cancer chemotherapy trials. J Clin Oncol. 2010;28(19):e319-e320. doi: 10.1200/JCO.2010.28.9934..
3. . Graybill WS, Pothuri B, Chase DM, Monk BJ. Poly(ADP-ribose) polymerase (PARP) inhibitors as treatment versus maintenance in ovarian cancer. Gynecol Oncol. 2017;146(1):11-15. doi: 10.1016/j. ygyno.2017.03.508.