Tomorrow's Medicine

Novartis’s Kymriah: Harnessing Immune System Comes With Worry About Reining in Costs

FDA approval of the CAR T-cell therapy for leukemia could usher in an era of genetically engineered, individually tailored immunotherapies. But tap those brakes. Long-term results are in short supply—and there’s that $475,000 price tag.

Thomas Morrow and Managed Care editors

For a parent, hearing the word leukemia may stir up one of the deepest forms of dread. It remains one of the most common types of childhood cancer, although leukemia can affect people of all ages. The name leukemia comes from the Greek leukos for white and heima for blood. The disease presents with a dramatic production of abnormal white blood cells. If not successfully treated, that uncontrolled production of white blood cells can cause death.

Leukemia is divided by the onset—acute or chronic—and by the type of cell—myeloid or lymphocytic. Thus, there are four primary types: acute and chronic myeloid (or myelogenous) leukemia (AML and CML) and acute and chronic lymphocytic (or lymphoblastic) leukemia (ALL and CLL). The lymphocytic group is further divided into the type of lymphocyte involved, T or B lymphocytes.

B-cell acute lymphoblastic leukemia (B-cell ALL) is the most common malignancy among children in the United States. Each year, about 5,000 American children and teens are diagnosed with B-cell ALL. Current therapy consists of multiagent regimens, and the cure rate exceeds 85%. But approximately 15% of children and young adults with ALL will relapse. In addition, between 2% and 3% of patients will fail induction therapy, a condition termed refractory. Collectively, refractory and relapsed B-cell ALL (r/r B-cell ALL) remains one of the leading causes of cancer death in children.

Treatment options for the estimated 600–700 American pediatric and young adult patients with r/r B-cell ALL each year include attempting reinduction with chemotherapy, targeted therapies, and stem cell transplants. But overall survival is dismal, measured in months.

That may all change very soon.

Not in Kansas anymore

In July, the Oncologic Drugs Advisory Committee of the FDA unanimously recommended approval of Novartis’s new gene therapy for ALL, which during its development was called tisagenlecleucel, or sometimes by its investigational moniker, CTL019. On August 30, the FDA approved tisagenlecleucel, which Novartis is marketing as Kymriah. Hurricane Harvey dominated the news that week, but the Kymriah approval was right up there in the news cycle.

The FDA billed it as the first gene therapy in the United States, which led to some debate about semantics and exactly what should be deemed gene therapy.

High drug prices are not news these days. Still, there was a lot of reaction when Novartis announced that it was pricing Kymriah at $475,000. Wall Street investors were dismayed because some analyses indicated that a price of $750,000 might be justified. But consumer groups and others didn’t let Novartis off the hook. “While Novartis’ decision to set a price at $475,000 per treatment may be seen by some as restraint, we believe it is excessive,” wrote David Mitchell, founder of Patients for Affordable Drugs, a consumer group that is campaigning against high drug prices. “Novartis should not get credit for bringing a $475,000 drug to market and claiming they could have charged people a lot more.”

Trying to describe this drug reminds me of when Dorothy in The Wizard of Oz says, “Toto, I’ve a feeling we’re not in Kansas anymore!” So let me try to explain our new location. And a disclaimer: This short article does not do justice to the complexity of this new world we are now entering. For a more complete understanding, I refer the readers to the FDA documents, which number in the hundreds of pages.

In those documents, Kymriah is described as “an autologous, immunocellular cancer therapy that involves reprogramming a patient’s own T cells with a transgene encoding a CAR to identify and eliminate CD19-expressing malignant and nonmalignant cells.” (Those documents don’t use the brand name but I am going to use Kymriah for simplicity’s sake.)

Got that?

No? OK, a few words of explanation.

While Kymriah is a new therapy, it’s not a medication in the way we typically think about medications. It’s a genetically modified white T cell that comes from a patient’s body. And it is “transgenic” (meaning it has genes from other organisms) and contains a new inserted antibody fragment called CAR that is programmed to stick—think Velcro—to a specific receptor on the patient’s cancerous B cells and kill them.

The genetic modification is why by some criteria this is gene therapy, although many reserve that term for therapy designed to replace or alter a gene. And this is one of the exciting new CAR-T therapies that you’ve probably been reading about.

CAR is an abbreviation for chimeric antigen receptor. This molecule comprises a single-chain antibody fragment, sourced from mice, that recognizes the CD19 receptor and is fused to two intracellular signaling domains, CD3-zeta and 4-1BB. The CD19 receptor is one of a family of “cellular differentiating” receptors located on specific types of white blood cells. The 19 in CD19 refers to a specific member of the CD family that is only found on cells of the B cell lineage. The CD3-zeta component is critical for activating T cells into performing anti-tumor actions. The 4-1BB component enhances the “expansion” (a huge increase in number) and, importantly, the persistence of this newly genetically engineered T cell. These cells reproduce in the body to engage and eliminate any CD19-expressing cells. They exhibit immunological endurance and produce long-lasting remission.

The manufacturing process for Kymriah is necessarily personalized because it starts with the patient’s own T cells. Patients have blood drawn in the same way that blood is drawn for a blood donation. Leukapheresis separates white blood cells (specifically the monocytes) from the red cells and serum, which is returned to the patient. The cells are frozen and shipped to a special Novartis facility in New Jersey for processing. A dedicated team, which works on just one project at a time, handles each patient’s cells.

After thawing, cells undergo a “cleansing” to remove all unneeded cells—such as monocytes and B-lineage lymphoblasts—that are detrimental to CAR transduction and growth. T cells are then activated in the lab and exposed to a “self-inactivating minimal lentiviral vector” that has the CD19 CAR transgene on board. The lentiviral vector is a modified HIV-1 virus—yes, the genome of the virus that causes AIDS. The needed CAR genetic material is introduced into the T cells and is incorporated into the genome of these cells in a process called transduction.

The transduced cells are grown in the special Novartis facility, washed and cryopreserved. After some testing, they are shipped to the site of care where the “drug” is infused into the patient in a weight-based dosing formula. This whole process can take weeks, and patients stay on standard treatment while they are waiting.

Documents filed with the FDA oncological committee went into detail about the HIV-1– sourced lentiviral vector. Obviously, great care was taken in the selection, design, and testing of the vector and the resultant vector-transduced cells. The researchers paid particular attention to the risk that the vector might lead to a secondary cancer or produce a pathogenic virus. Earlier forms of vectors used to transfer genes in experimental gene therapy demonstrated the ability, over time, to become capable of replication. Later generations of retroviral vectors were designed to prevent that from happening by selectively removing genetic material. Scientists start with the HIV-1 genome and remove all nonessential HIV-1 sequences such as those that produce the protein envelope and the accessory proteins needed for replication. The genes are then transferred to the patient’s T cells in small packets that are unlikely to recombine to “create” a viable virus.

Kymriah was first studied in a phase 1–2a trial to assess long-term persistence, in vivo proliferation, anti-tumor activity, and safety in patients with r/r and incurable B-cell malignancies. Promising results in this trial resulted in two trials, B2205J, a single-arm phase 2 trial, and B2202, (the pivotal trial), an international, multicenter, single-arm, open-label, phase 2 trial. The patients were between the ages of 3 and 21 at the time of diagnosis. Both B2205J and B2202 had nearly identical study designs and similar patient profiles in the enrollees. Cumulatively, all three trials included 147 treated patients.

Results were astounding, given the failure of conventional therapy and seriousness of the illness in this group of patients. Overall response rates in the three trials ranged from 69% to 95% and complete response rates ranged from 60% to 89%. The estimated relapse-free rate among responders at Month 6 was 75.4%. Median overall survival was available for only one trial at the time of the filing, but at 33 months it was impressive compared with other treatments.

The adverse events included cytokine-release syndrome (a class effect related to the mechanism of action); neurological toxicities—including aphasia, tremor, seizures, confusion, and encephalopathy—and a number of others that are typical of cancer treatment (tumor lysis syndrome, febrile neutropenia, prolonged B cell aplasia). Of note: No adverse events associated with the lentiviral vector were observed.

Who is going to pay?

Kymriah is nothing if not complex. Its development brings together the most cutting-edge technology of our time.

The results are stellar, but remember this is still not considered a cure. Kymriah was approved by the committee based on phase 1 and 2 trials, for which long-term follow-up is still required to determine medium overall survival because data were not available at the time of the application. Not enough people had died to reach a 50% level—good news, but it reflects how desperate the need is in this disease.

The price of Kymriah is going to be an issue, although it probably won’t be long before another, even pricier drug is approved. Relatively few people have r/r B-cell ALL, so payer budgets can probably absorb the cost of Kymriah without completely collapsing. But the CAR-T approach can be applied to other receptors, and researchers are busy investigating how to use it as a treatment for other cancers. The hope is that this is just the beginning of the CAR-T era in oncology.

Novartis is pursuing some value-based arrangements that may help blunt the effect of the high price of Kymriah—arrangements that could set important precedents for other drugs and other companies. So, for example, Novartis has approached CMS and some insurers about an outcomes-based approach that would limit charges to patients who respond to the drug in the first month. And Novartis may be moving toward charging different prices for different indications.

Still, it is hard not to be nervous. Are we on a collision course with therapies based on virtuoso 21st century science getting hopelessly ensnared in the messy reality of how drugs are priced and paid for?

Somewhere over the rainbow there may be a solution.

Thomas Morrow, MD, is a retired physician executive and part-time consultant. He has been the founding medical director of five HMOs and a disease management company, a medical director at Genentech, and president of the National Association of Managed Care Physicians. You can contact him at

Note: An earlier version of this story was first published on the Managed Care website on Aug. 31, 2017.

A blueprint for high-volume, high-quality lung cancer screening that is detecting cancer earlier—and helping to save lives

Clinical Brief

Multiple Sclerosis: New Perspectives on the Patient Journey–2019 Update
Summary of an Actuarial Analysis and Report