Surrogate End Points, Drug Costs Complicate Oncology Formularies

Accelerated approvals, ultra-high-cost therapies, and surrogate end points are straining the formulary decisions of comprehensive cancer centers and community oncology practices alike, according to pharmacy and physician leaders who spoke on a panel at The American Journal of Managed Care^®’s annual Patient-Centered Oncology Care® conference, held in Nashville, Tennessee, September 24-25, 2026.

The panel, moderated by Lalan Wilfong, MD, chief medical officer of Navista, described a pharmacy and therapeutics (P&T) landscape where an FDA approval no longer settles the question of whether a drug reaches a patient.

How Should P&T Committees Weigh Accelerated Approvals?

The panelists returned repeatedly to the ethical weight of accelerated approvals granted on small trial populations. Scott Soefje, PharmD, MBA, BCOP, FCCP, FHOPA, director of pharmacy cancer care at Mayo Clinic, noted that roughly 20% of accelerated approvals are ultimately withdrawn after their indications fail to hold up,¹ a track record that complicates every P&T committee review.

“Are we treating patients with a drug that has no effect?” he asked, framing the dilemma that now shapes P&T committee deliberations. “Do we approve all accelerated approvals that come through? We have to look deeper into it. We have to look at the data. It just creates this dilemma. I don't think we know the answer.”

Wilfong said the uncertainty is compounded by trial populations that increasingly look nothing like the patients whom community oncologists actually treat, particularly for rare indications and newer modalities such as bispecific antibodies and chimeric antigen receptor (CAR) T-cell therapies. Mohit Narang, MD, hematologist-oncologist and managing partner at Maryland Oncology Hematology, added that testing gaps widen the mismatch further: next-generation sequencing rates for non-small cell lung cancer still hover at only 50% to 60% in his practice, even as targeted therapies depend on that data to be used appropriately.

Even so, demand for newly approved drugs rarely waits for that evidence to mature. Jeanine Ksar, PharmD, BCOP, FCS, clinical oncology pharmacist at Florida Cancer Specialists & Research Institute, said patients often arrive already asking for a specific therapy because they have seen it covered in the news or added to clinical guidelines.

“You want to make it available to patients,” she said, while stressing that her team still weighs safety and efficacy data, National Comprehensive Cancer Network guideline status, and competing options before adding a drug to the formulary.

High-Cost Drug Committees

The price tags attached to newer therapies are forcing institutions to build financial infrastructure that did not exist a decade ago. Soefje described Mayo Clinic’s separate high-cost drug committee, distinct from P&T, that evaluates whether a practice has enough patient volume to justify carrying a therapy that can run into the hundreds of thousands or millions of dollars per dose.

He compared the cash flow burden with taking out a mortgage. The practices have to pay $2 million upfront for a drug, then it may take 2 weeks to treat the patient, and finally it could take 90 days for the payer to reimburse.

“How many people can hold a $2 million charge for 2 or 3 or 4 months?” Soefje asked. “That’s a tough thing, particularly when you start getting into the small practices.”

Mitchell Blewett, PharmD, MS, BCOP, BCPS, director of oncology pharmacy at Cleveland Clinic, said Cleveland Clinic has stood up a similar committee and now meets weekly with finance and supply chain teams to forecast which cellular and gene therapies will hit the budget in the coming year.

Both panelists noted that reimbursement rarely keeps pace with cost. Soefje said payers are often unwilling to pay even a 6% markup on drug administration charges once a therapy’s price reaches into the millions, calling the margins on products such as CAR T-cell therapy “razor-thin.”

When a Drug Is Approved on a Surrogate End Point

Surrogate end points added another layer of disagreement.² Soefje said his willingness to accept overall response rate as sufficient evidence depends heavily on disease state: in hematologic malignancies, response can reasonably predict cure, but in many solid tumors, response rate does not reliably translate to progression-free or overall survival. That leaves committees weighing where a drug fits into an already crowded treatment landscape.

“It’s real hard for us to say no to an FDA-approved drug at Mayo Clinic… Are you better, are you safer, are you cheaper? Do we need 13 PD-1 inhibitors?” Soefje said.

Narang said multiple myeloma has a clearer path toward using minimal residual disease as a surrogate,³ but he questioned how consistently that marker is being tested in community settings versus academic centers.

How Real-World Evidence Fills the Head-to-Head Data Gap

With fewer cooperative-group trials directly comparing competing drugs in the same class, Blewett said real-world evidence and matching-adjusted indirect comparisons (MAICs) are increasingly filling the vacuum, though cautiously. He said MAICs are time-consuming to evaluate and prone to bias, particularly when competing manufacturers present dueling analyses claiming superiority for their own product.

“The intent is good [because] often times we don't have the direct comparison—the head-to-head comparison—that we really need to make a decision and evaluate different options,” Blewett said. “That being said, I take it with a grain of salt. It's not something that I would have sway a decision.”

Blewett added that his team instead layers manufacturer-supplied financial and outcomes tools with its own institutional data before making a call.

Ksar said her organization relies on an internal real-world evidence database to evaluate drug sequencing questions that published trials do not directly answer, cross-checking outside data against its own patient outcomes. Soefje argued that well-designed real-world studies are likely to become a permanent substitute for the head-to-head trials cooperative groups can no longer afford to run. However, he cautioned that real-world evidence varies widely in quality and needs the same scrutiny applied to clinical trial data.

The panelists closed by returning to formulary decisions that ultimately come down to the patient in the room: once a drug clears committee review, Wilfong said, the remaining work is shared decision-making between physician and patient.

“Our core value is what's in the best interest of the patient,” Soefje said. “I think what we need to get better at is…that shared decision-making: present the data, present the side effects, present how you take this drug, and then work with the patient to identify what's the best drug for that particular patient.”

References

1. McCormick B. Concerns persist over reliance on surrogate end points in FDA accelerated approvals. AJMC^®. July 24, 2025. Accessed September 24, 2026. https://www.ajmc.com/view/concerns-persist-over-reliance-on-surrogate-endpoints-in-fda-accelerated-approvals

2. Parikh RB, Hubbard RA, Wang E, et al. Exposure to US cancer drugs with lack of confirmed benefit after US Food and Drug Administration accelerated approval. JAMA Oncol. 2023;9(4):567-569. doi:10.1001/jamaoncol.2022.77703

3. Shaw ML. FDA draft guidance proposes CR, MRD to accelerate MM drug approval. Am J Manag Care. 2026;32(Spec 2):SP91.