Medicare's CAR-T Panel Underscores Hurdles For PROs – And FDA/CMS Collaboration

MEDCAC meeting demonstrates complexity of patient-reported outcomes and place for PROs in regulatory decision-making compared to coverage.

The US Medicare Evidence Development & Coverage Advisory Committee's review of CAR-T therapies had the potential to set up as a coming-out party for patient-reported outcomes (PROs) as a driver of development and coverage.

Instead, the event provided a sobering look at where PROs are on the regulatory and reimbursement continuum.

Technically speaking, the meeting was the next step of a national coverage analysis process triggered by [UnitedHealthCare]for the two currently approved CAR-Ts: Novartis AG’s Kymriah (tisagenlecleucel) and Gilead Sciences Inc./Kite Pharma Inc.’s Yescarta (axicabtagene ciloleucel).

The committee endorsed a suite of patient-reported outcomes assessments that had the common thread of historic use in clinical trials. (Also see "Medicare Advisers Endorse Patient Reported Outcomes For CAR-T With Caveats " - Pink Sheet, 3 Sep, 2018 and “CAR-T Patients’ Ongoing Needs After Treatment Highlighted At MedCAC Meeting” – Pink Sheet, 3 Sep, 2018)

However, the seemingly straight-forward end result of the panel votes belied the confusion, complexity, and challenges of how to use PRO and clinical outcomes assessment (COA) tools when it comes to coverage of novel therapies.

A Novel FDA/CMS Collaboration

Prior to the meeting, one name stood out on the agenda: FDA’s Paul Kluetz, Oncology Center of Excellence Associate Director of Patient Outcomes, as a guest presenter. It’s the first time in recent memory a senior FDA official – or any FDA official – had been asked to make a formal presentation to the MEDCAC on a drug-related issue.

For example, the 2010 MEDCAC review of Dendreon Corp.’s Provenge featured a presenter from the National Cancer Institute, but not FDA – arguably the agency with the deepest understanding of the therapy considering it reviewed and approved the application containing the individual case reports.

Even the landmark 2011 MEDCAC meeting on the erythropoiesis-stimulating agents (ESAs) in patients with chronic kidney disease failed to showcase a single FDA presenter despite the agency’s knowledge and regulatory history with the ESAs.

Similarly, a more recent MEDCAC in 2016 on recommendations for defining treatment-resistant depression – more of a device issue but with obvious overlap with drug therapy – didn't have an FDA presenter.

So when Kluetz’ name appeared on the agenda, it signified a novel FDA/CMS collaboration of sorts given the previous absenses.

FDA's Big Picture

Kluetz has been a leader at FDA in trying to more finely tune PROs/COAs so they can be more usable for regulatory decision-making.

He most recently led the agency’s third clinical outcome assessments in cancer trials workshop June 22. That workshop demonstrated just how deep in the weeds the agency has gone to in order to better understand and design PROs and featured a range of perspectives on instruments, libraries, fake case studies and trial designs.

At the MEDCAC, Kleutz’ presentation attempted to take a step back and look at the big picture of the PRO world. Kluetz started with terminology since the PRO world is one unto itself.

Kluetz walked through what a well-defined fit-for-purpose PRO instrument: appropriate for its intended use; validly and reliably measures clinically relevant concepts that are important to patients; and can be communicated in labeling in a way that is accurate, interpretable and not misleading.

Kluetz used a physical function scale as an example of a well-defined PRO. He has previously cited safety and tolerability as appropriate PRO when thinking about a statistically measurable endpoint.

The Emoji Index

Despite Kluetz’ best efforts, MEDCAC members were mostly confused over how to rank and evaluate the PRO tools they were being asked to vote on.

If it weren’t for an effective presentation by Oncologist Ethan Basch (University of North Carolina) – who actually used smiley face and sad face emojis in his slide deck to show which PRO tools he would endorse in the voting – the panel likely would have been completely.

Remarkably, the four tools with smiley face emojis on Basch’s slide at the end of his presentation were the exact four MEDCAC endorsed: PRO-CTCAE, MDASI, EORTC-QLQ-C30, and PROMIS.

Joseph Cheng (University of Cincinnati) said that many of the PROs were “too onerous and just start out” with PROMIS.

James Yang (NCI) was skeptical of mandating PRO use: “I don’t know how I would apply it in a uniform consistent fashion.”

PROs "Not Quite Ready"

The CAR-T sponsors are heavily invested in PROs and incorporating them into clinical trial designs; however, the sponsors are unified in their opposition of them being a requirement of coverage.

“The science behind where these instruments are most appropriate is still evolving,” Kite VP-Clinical Development William Go said. “We feel very strongly that it’s best served in a randomized Phase III study. We feel that PROs are not quite ready for real-world coverage decisions at this time. We do not believe that PRO data collection should be linked to coverage for these transformative therapies at this time.”

Novartis’ Ilia Ferrusi, associate director, CAR-T therapy, US oncology, agreed. “Mandating patient reported outcomes data collection for CAR-T is unnecessary, impractical, and imposes a significant burden on providers and patients outside of clinical studies.”

Celgene/Juno’s Karen Chung, senior director, health economics and outcomes research, said “we strongly support the incorporation of the patient voice into clinical trials, but firmly believe PROs should not be a condition of coverage.”

CMS Likely Needs More Time

At this point, it’s unclear how CMS is going to use the MEDCAC to inform its national coverage decision on the CAR-Ts.

“I don’t think this is the end of this conversation,” CMS’ Tamara Syrek Jensen, Director of Coverage and Analysis Group in the Center for Clinical Standards and Quality, said of the implications of using PROs in coverage decisions.

A reasonable interpretation of the meeting outcome is that CMS will need more time and data in order to the most appropriate and effective PRO tools for use in assessment of the Medicare population. While the MEDCAC provided some guidance, the conviction of the committee seemed relatively low.

The proposed NCD is due by Feb. 16, 2019.

From the editors of the RPM Report