[Corrected, 1/4/19, 3:55pm ET. See below.] A generation ago, cancer treatments made from a patient’s own living immune cells would have been science fiction. Now they’re here. The first two products, approved in 2017 and known in shorthand as CAR-T, have brought some people with otherwise untreatable blood cancers back from the brink of death.
“Often these patients have survival measured in weeks to months,” says David Maloney, medical director of the Bezos Family Immunotherapy Clinic in Seattle, which at any given time is treating more than 30 patients with both experimental and approved CAR-T products.
The experimental results that led to approval for each product were remarkable. But outside the cautious, tightly controlled bubble of clinical studies, the products and their owners have had different experiences in the real world.
One CAR-T, axicabtagene ciloleucel (Yescarta) from Gilead Sciences (NASDAQ: GILD), has provided surprising real-world results so far that hew closer to the successful clinical studies than doctors typically expect. The safety record in particular is “reassuring,” says Caron Jacobson, who treats leukemia and lymphoma patients at Dana-Farber Cancer Institute in Boston, one of the top country’s top CAR-T centers. Side effects from the aggressive cancer-killing cells have been a major concern for doctors.
The other approved CAR-T, tisagenlecleucel (Kymriah) from Novartis (NYSE: NVS), has hit major snags. While its clinical results were also strong, Kymriah has been used far less in the real world. Novartis has had problems manufacturing the complicated product, which, like Yescarta, requires a patient’s T cells to be extracted, genetically modified in a lab, nurtured so the cell population expands, and shipped back to the patient.
(CAR-T stands for chimeric antigen receptor T cell, a reference to the genetic engineering that programs the killer cells to seek out specific proteins on tumors. Yescarta and Kymriah are autologous CAR-T products, which take about three weeks to produce and administer to the patient.)
Kymriah was first approved in August 2017 for pediatric acute lymphoblastic leukemia (ALL), a rare blood cancer. Novartis officials said at the time they expected 600 eligible patients a year. In May 2018, Kymriah received approval for adults with the most common form of non-Hodgkin lymphoma (NHL).
Yescarta was approved in October 2017 for several types of adult NHL.
Both are allowed only after other treatments, including chemotherapy and bone marrow transplant, have been exhausted or ruled out. And both come with price tags that have raised serious concerns about affordability, from health policy experts to the FDA commissioner.
Kymriah has a one-time price of $475,000 for ALL, with a refund if it doesn’t work after one month, and $373,000 for NHL (no refund). Gilead has priced Yescarta at $373,000, with no refund. [Correction: The price of Kymriah for NHL is $373,000, not $375,000.]
(A third CAR-T for NHL could be on the way. If the megamerger of Bristol-Myers Squibb (NYSE: BMY) and Celgene (NASDAQ: CELG) is consummated, Bristol’s large experimental CAR-T portfolio will include JCAR017, which was developed by Celgene division Juno Therapeutics. It remains to be seen if Bristol will push JCAR017 aggressively in NHL after hints last month that Celgene’s enthusiasm might have waned.)
Maloney of the Bezos Clinic says Kymriah’s experimental record suggests it could have less intense side effects, thanks to a difference in the engineering. But he prefers Yescarta for lymphoma patients for a simple reason: “Novartis is still struggling to produce Kymriah. I can’t get it from them.”
At University of Texas MD Anderson Cancer Center in Houston, one of the top CAR-T treatment centers in the country, the number of patients treated with Kymriah as of late November were “in the single digits,” says Sattva Neelapu, a blood cancer specialist. At Dana-Farber in Boston, another major CAR-T center, the total is two.
Those figures are reflected in the financial statements of the product’s owners. Yescarta garnered $183 million in revenue the first three quarters of 2018 for Gilead, which bought Yescarta’s developer Kite Pharma for nearly $12 billion a few weeks before the product’s October 2017 approval.
Kymriah tallied $48 million in revenue through the first three quarters of 2018. (The timing of its approvals, first in ALL, a tiny patient population, then in the larger NHL, is one factor but not a full explanation.)
Neither company would give an exact count of their real-world patients. But it’s safe to say a couple hundred if not more have received Yescarta in a real-world setting, enough for oncologists to start analyzing its performance. Two studies were presented at the American Society of Hematology’s annual meeting last month. Patients receiving Yescarta commercially fared a bit worse than patients in the 101-patient study known as ZUMA-1, which Kite Pharma used to gain FDA approval.
That may not sound praiseworthy at first. But once a drug is available commercially, a wider swath of patients, often closer to death or with other layers of health problems, can receive treatment; conditions are less predictable, less tightly controlled.
In Yescarta’s case, the performance gap between experimental conditions and the real world has been narrow so far. “There is a drop-off of results in every oncology drug, but here, I am surprised,” says Neelapu of MD Anderson, who is a co-leader of the extended ZUMA-1 study.
Neelapu is particularly pleased that the real-world safety numbers across two studies and 188 patients were similar to ZUMA-1: “I expected higher toxicity because of sicker patients.”
Across the Bridge
One potential factor in the better-than-expected results, says Neelapu, is short-term “bridging” therapy for patients who have only weeks or months to live. CAR-T production takes three weeks; some patients might not survive that long, so doctors give them a burst of radiation or other treatment as a temporary bridge to CAR-T. The quick hit might be shrinking tumors enough to decrease the CAR-T side effects, which can be amplified by the amount of cancer in a patient’s body. Think of the killer T cells as sharks at a feeding frenzy, and the cancer is the food. The less there is to eat, the less collateral damage during the attack.
There are still side effects, to be sure. But patients can only be treated with CAR-T in centers where staff receive special training and often have experience with bone marrow transplants, which are also fraught with complications. They have steroids and other drugs at hand to tamp down an immune system threatening to run amok, and the intensive care unit is at the ready.
In ZUMA-1, it’s now been more than two years since many patients received their one-time treatment. In an update last month, Yescarta helped shrink the tumors of 83 percent of patients with 58 percent showing no sign of cancer. (Median follow-up of 27.1 months.) Some observers say it takes three to five years of remission to start using the “C” word—cure—but these types of results, two years after a one-time treatment, are remarkable.
In the real-world reports released last month, patients were not followed nearly as long. But the results were promising:
–For 112 patients treated across 17 medical centers, including MD Anderson and the University of California, San Francisco, the one-month outcomes were 79 percent showing some response; 50 percent were showing no sign of cancer. In the same study, 163 patients were available for safety evaluation. There are two main types of serious side effects with CAR-T, which receive a grade 3 or higher: an overheated immune system response known as cytokine release syndrome (CRS); and neurological problems. CRS occurred in 7 percent of patients. Neurological problems occurred in 31 percent. Those numbers were very similar to those seen in ZUMA-1 (11 percent and 32 percent).
—For 76 patients treated across a different set of centers, including Dana-Farber Cancer Institute in Boston and the Seattle Cancer Care Alliance, 57 percent of patients had some type of response, with 36 percent showing no sign of cancer. The median follow-up after receiving the CAR-T cells was four months. Thirteen percent of patients had grade 3 or higher CRS, and 38 percent had grade 3 or high neurological problems. There were no 4-month data from ZUMA-1, but at the six-month check-in two years ago, about 35 percent of patients had seen their cancers essentially disappear.
More than 60 percent of the patients in the study would not have met the bar to enter ZUMA-1, at least in part because they were too sick. The similar results “may mean that the data from ZUMA-1 is an accurate portrayal of the safety of Yescarta,” regardless of a patient’s status, said Jacobson, the first author of the study and medical director of Dana-Farber’s immune cell therapy program.
CAR-T For Free
Kymriah, meanwhile, has been dogged by manufacturing troubles. The main problem: After genetic engineering outside the patient, not enough cells have been converted into active cancer killers. “When final product does not meet specification, it is usually due to lower than expected percentage of viable cells,” says Novartis spokesman Eric Althoff. (Manufacturing problems also occurred with nearly 8 percent of enrolled patients in Novartis’s two key clinical trials.)
The firm’s top cell and gene therapy official told Biopharmadive last month that many patients were receiving Kymriah for free, because Novartis can’t charge for a product that doesn’t meet the specifications on its commercial label.
On an earnings call late last year Novartis executives tried to put the situation in a positive light. “We’re actually above our expectations given the manufacturing challenges that we’ve had,” said Elizabeth Barrett, the former head of the Novartis oncology division. “Centers have continued to order Kymriah despite this.”
On the same call, Novartis CEO Vas Narasimhan said Kymriah’s safety record has encouraged Novartis to expand manufacturing in France, Switzerland, and China. (The real-world evidence for Kymriah is only anecdotal. No studies have yet emerged. Long-term results from Kymriah’s clinical study in NHL were released late last year.)
Narasimhan was also optimistic that CAR-T could treat patients in earlier stages of disease, not just end-of-the-line cases. “We really see this as a long-term play, and we are quite confident in the profile of the medicine,” he said.
For others to feel the same confidence, however, more data are necessary, and not just for this first generation of CAR-T products. “Watch the safety data of next-generation CAR-T closely,” says Morris Paterson of Group H, a biopharma consulting group. Several companies, including Novartis and Gilead, have experimental CAR-Ts in development for blood cancers and more. “If they have a better safety profile, using them in a broader setting becomes more feasible.”
More CAR-T, Cheaper CAR-T?
One of Novartis’s biggest critics also wants to see the cutting-edge medicine become an option for patients in earlier stages of disease. David Mitchell, a public-relations executive and multiple myeloma patient, is the founder of Patients for Affordable Drugs, a drug-price watchdog. He slammed Novartis when it announced Kymriah’s $475,000 price tag, noting how U.S. taxpayers had shouldered more than $200 million to fund the research that led to CAR-T.
The federal agency overseeing the Medicare and Medicaid insurance programs has been weighing how much to pay for CAR-T—a final decision could come this spring. Mitchell’s group has suggested a pay-for-performance scheme that rewards drug makers with milestone payments based on the patient’s condition one, six, and 18 months after infusion. Mitchell has a personal stake in his advocacy. Several companies are racing to develop CAR-T products for multiple myeloma. (Celgene is in the lead—another reason to watch the proposed acquisition closely.)
The cancer, which affects cells in the bone marrow, almost always progresses, even after long remissions. Patients cycle through several treatment regimens. If CAR-T becomes safer to give to patients who aren’t nearly as sick, and it proves to be durable, Mitchell argues that higher prices might be justified if they replace several levels of regimens. “I don’t want to go through two or three lines of treatment and millions of dollars,” he says. His current regimen cost $450,000 the first year and $300,000 each year thereafter. “The difference between ‘my cancer is going to progress, I’m going to run out of drugs and die,’ and ‘I have at least 10 more years because of CAR-T,’ is a big deal,” Mitchell says.
Even with last month’s release of promising real-world results for Yescarta, no one knows yet if it or Kymriah can deliver the same longterm remissions that they have in leukemia and lymphoma clinical studies. Until then, the argument for moving them and perhaps other CAR-T products into earlier-stage cancer will be tough to make at their current prices, and one of the most promising medical advances in decades could end up solely as a last-ditch treatment for handfuls of patients. Good for those patients, but disappointing for those expecting CAR-T to make a major dent in cancer.
Image of a T cell attacking cancer courtesy of the National Cancer Institute/Duncan Comprehensive Cancer Center at Baylor College of Medicine.