When the Label Lags the Science: Using 505(o)(4) to Compel Labeling Changes

When perusing regulations.gov, as bloggers searching for content are wont to do, we ran across an interesting Citizen Petition from June that grabbed our attention.  Indeed, a June 11, 2026 Petition from Marc R. Matrana, M.D.—System Medical Director of Precision Medicine at the Ochsner MD Anderson Cancer Center in New Orleans and a longtime advocate for precision-oncology legislation in Louisiana—caught our eye because of the novel use of FDC Act § 505(o)(4).  Typically reserved for when the Agency learns of new safety issues from adverse events or publications, FDC Act § 505(o)(4) allows FDA to request that a sponsor update its product labeling when the Agency “becomes aware of new information, including any new safety information or information related to reduced effectiveness, that [FDA] determines should be included in the labeling of the drug.”  In other words, FDC Act § 505(o)(4) instructs FDA to initiate labeling changes once it becomes aware of new safety information, which can include effectiveness (FDA has interpreted the “effectiveness” issue to mean that if a drug’s ineffectiveness leads to death, risk of death, hospitalization, or incapacitation, it can be a basis for a mandated labeling change.).  This Petition uses FDC Act § 505(o)(4) as a lever to force updated scientific information into labeling.  It’s an interesting use of the regulatory procedure.

The Petition asks FDA to direct a labeling change for a group of immunotherapies used to treat NSCLC and adopt a “Limitation of Use” to exclude patients whose tumors carry ROS1 or RET rearrangements so that these patients are redirected to targeted TKIs.  Rather than target a single product, the Petition asks FDA to direct labeling changes across six major immunotherapies (KEYTRUDA, OPDIVO, YERVOY, LIBTAYO, TECENTRIQ, and IMFINZI) simultaneously.  Thus, this Petition asks FDA to use its authority to change labeling for an entire swath of products in one fell swoop, reshaping an entire drug class at once.  If FDA were to grant it, that would demonstrate the statute can be used to make coordinated, class-wide labeling revisions based on evolving evidence—a meaningful precedent for how the agency polices approved oncology labels.

That said, the underlying premise—that oncology labeling can go stale as the evidence base evolves, and that FDA should do something about it—isn’t new to the agency.  FDA’s Oncology Center of Excellence already runs Project Renewal, a standing initiative to revisit older oncology drug labels against current science; it has produced updated labeling for drugs like Xeloda, Temodar, and fludarabine phosphate.  What’s novel here isn’t the idea, but the mechanism: rather than a collaborative FDA-sponsor label update, the Petition asks FDA to use its compulsory §505(o)(4) authority to force the issue across six different sponsors’ products simultaneously.

The Petition itself makes an interesting argument.  It cites recent revisions to NCCN Clinical Practice Guidelines, which no longer recommend immunotherapy as a first-line option for patients with ROS1 or RET fusions, as “new safety information” within the meaning of section 505(o)(4).  The Petition implies that the peer-reviewed data and postmarket observational evidence underlying the NCCN Clinical Practice Guidelines render them a synthesis of newly available scientific information that FDA must consider.  This new evidence, published since the named immunotherapies were approved, shows that using them as first-line therapy in ROS1- and RET-rearranged patients introduces real safety risks, including reduced effectiveness, because immunotherapies do not target the genetic driver of the disease. By failing to warn clinicians that ICIs are inappropriate as first-line therapy for these populations, the Petition contends that the labeling creates a reasonable likelihood that clinicians will initiate a therapy that offers minimal benefit, delays access to effective TKIs, and increases hospitalization risk; the labeling is thus “misleading” absent a Limitation of Use for ROS1 and RET populations.

Underlying all of this is a practical reality: the uptake of biomarker testing is limited, resulting in serious complications for patients.  According to the Petition, roughly half of advanced NSCLC patients who could benefit from personalized treatment do not receive it, and even among patients who are tested and get timely results, a meaningful share still do not receive appropriate targeted therapy.  Because clinicians often rely on drug labeling to guide sequencing decisions, the Petition argues that labeling silence on first-line ICI use in ROS1 and RET patients contributes directly to inappropriate therapy selection.

If FDA acts, it would confirm that section 505(o)(4) can be used to align an entire drug class with updated guidelines and postmarket evidence.  If FDA declines, the response may reveal the statutory threshold for “new safety information.”  Either way, the filing underscores a larger point for anyone tracking precision oncology: keeping drug labeling in step with the science is increasingly a public-health imperative —an imperative FDA has already institutionalized through Project Renewal.