Optimizing FDA Submissions For Companion Diagnostics: Alternative Evaluation Pathways To Streamline Approval

By Janice Hogan and Shilpa Prem, Hogan Lovells

In part one of this two-part series, we examined the history of companion diagnostic approvals by FDA and identified trends to suggest better approaches to optimize the review process. Here, we will analyze new avenues sponsors may consider to further streamline the review of these important products through the assessment of various regulatory submission types and FDA guidance.

Although FDA has indicated that many companion diagnostics may require PMA approval, there has been evidence in recent years that FDA is willing to consider alternative pathways.

In 2012, FDA cleared a de novo submission for the FerriScan R2-MRI Analysis System, which is indicated to measure liver iron concentration to aid in the identification and monitoring of non-transfusion-dependent thalassemia patients receiving therapy with deferasirox. FDA created a generic product classification for this test, “liver iron concentration imaging companion diagnostic for deferasirox.”  The de novo submission was supported by data from the prospective study of deferasirox in this population, which included 166 subjects.

In assessing what factors may have permitted a de novo pathway for this submission — compared to a PMA pathway for many other companion diagnostics — it is notable that the Ferriscan R2-MRI test had been 510(k) cleared and marketed, for broader use in assessing liver ion concentrations, for nearly a decade prior to the companion diagnostics submission. Additionally, unlike many other companion products, the Ferriscan product is an image analysis software tool.

Although the risks of the test are generally similar to the risks of other companion diagnostics (i.e., erroneous results leading to incorrect decisions to provide or withhold deferasirox), the magnitude of these risks in the context of thalassemia may be lower and/or more manageable than, for example, companion diagnostics used with chemotherapeutics. 

FDA also noted that “the haematologist or gastroenterologist managing the patient always has the option to perform an invasive liver biopsy if they suspect the companion diagnostic does not accurately reflect the overall clinical condition of the patient.” Nonetheless, many of the risk mitigations applied to this test could also be applied to other types of companion diagnostics.

In 2015, ARUP Laboratories received two FDA approvals for companion diagnostics through the Humanitarian Device Exemption (HDE) pathway. Both of these companion diagnostics are intended to aid in the selection of patients for whom Gleevec (imatinib mesylate) treatment is being considered. One of the approvals was for a test that detects the KIT D816V mutation status from fresh bone marrow samples of patients with aggressive systemic mastocytosis. The other HDE approval was for a test intended to qualitatively detect the PDGFRB FISH gene in fresh bone marrow samples for patients with Myelodysplastic Syndrome/Myeloproliferative Disease (MDS/MPD). 

The HDE program is intended to benefit patients by treating or diagnosing a condition that affects a very small patient population (fewer than 4,000 in the United States, per year).  An HDE application is similar to a PMA but requires evidence of “safety and probable benefit,” which is a lower standard compared to the “safety and effectiveness” standard that applies to PMA applications.

Data requirements for these two HDE approvals reflect the rare nature of the genetic targets in each case. For example, for the PDGFRB gene detection assay, clinical data included an initial set of 11 previously collected specimens (obtained from 8 patients) positive for the PDGFRB rearrangement, and 97 patients negative for the PDGFRB rearrangement. The HDE approval was based on a data set that included 31 patients with MDS/MPD, including 7 patients with MDS/MPD from an open label phase II study of Gleevec in selected populations, and 24 patients from published reports or other studies. Similarly, for the D816V mutation test, the HDE approval was based on clinical data from five patients in an open-label phase II study of Gleevec, as well as 10 published case reports and case series with 23 additional patients. Both HDE approvals also were supported by substantial analytical testing.

Thus, use of the HDE pathway for two diagnostics intended for use in rare diseases appears to be the first use of the humanitarian device process for companion diagnostics. These examples illustrate a new potential pathway that could be leveraged for other companion diagnostics intended for use in rare diseases. 

Expediting the Approval of Companion Diagnostics

In April 2015, FDA issued its Expedited Access for Premarket Approval and De Novo Medical Devices Intended for Unmet Medical Need for Life Threatening or Irreversibly Debilitating Diseases or Conditions — Guidance for Industry and Food and Drug Administration Staff (EAP Guidance). In the guidance, FDA devoted significant attention to companion diagnostic products, due to their medical importance. Generally, a device qualifies for EAP designation if it satisfies the following criteria:

1.      The device is intended to treat or diagnose a life-threatening or irreversibly debilitating disease or condition; and

2.      The device meets at least one of the following criteria:

a.       No appropriate alternative treatment or means of diagnosis exists.

b.      The device represents a breakthrough technology that provides a clinically meaningful advantage over existing legally marketed technology.

c.       The device offers significant, clinically meaningful advantages over existing legally marketed alternatives.

d.      The availability of the device is in the best interest of patients; and

3.      The sponsor submits an acceptable draft Data Development Plan.

It should be noted that companion diagnostic devices do not “automatically” qualify for EAP designation, even if they are developed for use with drugs that are orphans, or designated for fast track or accelerated approval. However, FDA cites in the guidance examples of companion diagnostics that may qualify for the EAP program where their use is in the best interest of patients (e.g., to minimize serious side effects or target important therapies).

Importantly, FDA states in the guidance, with respect to the data required for companion diagnostics that qualify for EAP designation:

The types of evidence that might support PMA approval of companion diagnostics under the Expedited Access PMA would include intermediate and surrogate endpoints as well as clinical endpoints. The specifics will depend on the product and the study proposals provided by the sponsor. For example, if a drug is reviewed via the accelerated drug approval pathway based on a surrogate endpoint, the companion diagnostic may be considered for the EAP pathway. Because the safety and efficacy of a companion diagnostic is linked to that of the therapeutic product, the surrogate endpoint for the IVD will be that determined to be acceptable for the therapeutic product. This surrogate endpoint will be determined using processes already established by CDER or [Center for Biologics Evaluation and Research] CBER. However, CDRH encourages early discussion of surrogate endpoints in the context of the request for EAP designation during the Pre-Sub process.

Because surrogate endpoints historically have not been commonly used in medical device studies, this language recognizes the unique potential importance of surrogates presented by many companion diagnostic tests.

In the first year of the EAP program, 29 decisions were made on requests for designations, of which 17 were granted. The number of products granted EAP status includes both diagnostics and therapeutics.  It does not appear that any companion diagnostics have yet been able to leverage the EAP program, although experience to date is limited.  Because the EAP program remains, its impact on further reducing review times or otherwise streamlining the device approval process is as yet unknown, although preliminary experience to date is encouraging.

Conclusions

FDA has achieved faster review times for companion diagnostic PMA applications than almost any other type of class III product, with an average PMA review time in 2015 of less than 150 days.  Given the speed with which the FDA has reviewed these submissions, there may be limited potential for further reductions in review times within the PMA process.   

However, alternative procedures may offer the opportunity for further streamlining the  approval process for personalized medicine products. The Agency has shown flexibility and creativity in using approval mechanisms other than the PMA process — including de novo clearance and humanitarian device approval — in select cases, as well as supporting the use of the new EAP program for eligible companion diagnostic products. 

Assuming continued growth in the use of companion diagnostic products, additional exploration of the de novo pathway as an approval mechanism for at least some companion diagnostic products may help to further streamline product approval, reserving the PMA mechanism for those companion diagnostics that present the highest risk. Where companion diagnostics offer a tool to better target therapeutics, as an alternative to treatment with no adequate targeting, sponsors may be able to rationalize the use of special controls to allow de novo clearance.  The ability to leverage this mechanism will depend on the risks of erroneous test results associated with use or withholding of the target therapeutic.  Allowing use of the de novo pathway in appropriate cases would also allow sponsors to pursue future modifications in 510(k) submissions rather than more burdensome PMA supplements.

Pending further clarification of FDA’s emerging policy on laboratory developed tests, there may be an additional influx of companion lab tests for FDA review. Thus, there is a need for additional mechanisms to address the growing demand for FDA review of companion diagnostic technologies, consistent with the goals of the Personalized Medicine Initiative (PMI) launched by President Obama last January.

About The Authors

Janice Hogan is the co-director of Hogan Lovells' FDA/Medical Device practice. Janice focuses her practice primarily on the representation of medical device, pharmaceutical, and biological product manufacturers before the U.S. Food and Drug Administration (FDA).

A biomedical engineer, she held positions in marketing/marketing research for a major pharmaceutical manufacturer prior to becoming an attorney. She has authored articles regarding the medical device 510(k) review process, regulation of medical software, orphan drug regulation, use of foreign clinical data, the de novo review process, medical device products liability, and chapters of several textbooks related to medical device regulation.

Janice has served as an adjunct professor at the University of the Sciences in Philadelphia and as a guest lecturer at the Wharton School, Drexel University, and Stanford University on the development and regulation of medical products. She is also a frequent lecturer at FDA regulatory law symposia and conferences on topics related to premarket approval of medical products, combination products regulation, and product development. She has also served on the faculty of FDA's staff college training for new review staff.

Shilpa Prem is an associate at Hogan Lovells and represents medical device, pharmaceutical, and biological product manufacturers before the FDA. Shilpa has a biomedical engineering degree and has held various positions in large pharmaceutical and medical device companies.

Questions to the author may be directed to janice.hogan@hoganlovells.com.